From 2a8f9e2ba2b7e06f195cebcad6b048f26b494650 Mon Sep 17 00:00:00 2001
From: sion123 <450702724@qq.com>
Date: Sun, 16 Nov 2025 23:32:19 +0800
Subject: [PATCH 1/2] =?UTF-8?q?=E4=BC=98=E5=8C=96?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 .../module/tik/file/service/TikUserFileServiceImpl.java     | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/file/service/TikUserFileServiceImpl.java b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/file/service/TikUserFileServiceImpl.java
index f9490b48b7..6fddcada83 100644
--- a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/file/service/TikUserFileServiceImpl.java
+++ b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/file/service/TikUserFileServiceImpl.java
@@ -115,7 +115,7 @@ public class TikUserFileServiceImpl implements TikUserFileService {
 
         // ========== 第三阶段：保存数据库（在事务中，如果失败则删除OSS文件） ==========
         try {
-            return saveFileRecord(userId, file, fileCategory, fileUrl, filePath, coverBase64, baseDirectory);
+            return saveFileRecord(userId, file, fileCategory, fileUrl, filePath, coverBase64, baseDirectory, infraFileId);
         } catch (Exception e) {
             // 数据库保存失败，删除已上传的OSS文件
             log.error("[uploadFile][保存数据库失败，准备删除OSS文件，URL({})]", fileUrl, e);
@@ -129,7 +129,7 @@ public class TikUserFileServiceImpl implements TikUserFileService {
      */
     @Transactional(rollbackFor = Exception.class)
     public Long saveFileRecord(Long userId, MultipartFile file, String fileCategory,
-                               String fileUrl, String filePath, String coverBase64, String baseDirectory) {
+                               String fileUrl, String filePath, String coverBase64, String baseDirectory, Long infraFileId) {
         // 7. 处理视频封面（如果有前端传递的 base64 封面，先处理封面再插入主记录）
         String coverUrl = null;
         if (StrUtil.isNotBlank(coverBase64) && StrUtil.containsIgnoreCase(file.getContentType(), "video")) {
@@ -180,7 +180,7 @@ public class TikUserFileServiceImpl implements TikUserFileService {
         // 8. 创建文件记录（保存完整路径，包含封面URL和Base64）
         TikUserFileDO userFile = new TikUserFileDO()
                 .setUserId(userId)
-                .setFileId(null)  // 显式设置为null，file_id是可选的，用于关联infra_file表
+                .setFileId(infraFileId)  // 关联infra_file表，用于后续通过FileService管理文件
                 .setFileName(file.getOriginalFilename())  // 保存原始文件名，用于展示
                 .setFileType(file.getContentType())
                 .setFileCategory(fileCategory)

From 81f531b51b605fd37ea279805c49957f192a81b7 Mon Sep 17 00:00:00 2001
From: sion123 <450702724@qq.com>
Date: Tue, 18 Nov 2025 23:30:31 +0800
Subject: [PATCH 2/2] =?UTF-8?q?=E4=BC=98=E5=8C=96?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 .cursor/rules/backend.mdc                     |   23 +-
 CLAUDE.md                                     |  511 ++++++
 frontend/app/web-gold/src/api/mix.js          |   46 +
 frontend/app/web-gold/src/api/voice.js        |  110 ++
 .../src/components/ChatMessageRenderer.vue    |  155 +-
 frontend/app/web-gold/src/router/index.js     |    1 -
 frontend/app/web-gold/src/stores/voiceCopy.js |  174 +-
 .../app/web-gold/src/utils/video-cover.ts     |    2 +-
 frontend/app/web-gold/src/views/dh/Avatar.vue |   10 +-
 frontend/app/web-gold/src/views/dh/Video.vue  |  951 +++++-----
 .../app/web-gold/src/views/dh/VoiceCopy.vue   | 1542 ++++++-----------
 .../src/views/material/MaterialList.vue       |  169 +-
 .../app/web-gold/src/views/mix/MixEditor.vue  |   22 -
 .../file/core/client/s3/S3FileClient.java     |   10 +-
 yudao-module-tik/DESIGN.md                    |  282 ---
 yudao-module-tik/LOGIC_ANALYSIS.md            |   87 -
 yudao-module-tik/LOGIC_REVIEW.md              |  111 --
 yudao-module-tik/UPLOAD_STRATEGY.md           |   76 -
 .../module/tik/enmus/ErrorCodeConstants.java  |    8 +
 .../file/service/TikUserFileServiceImpl.java  |  168 +-
 .../service/TikFileTransCharacters.java       |   36 +-
 .../tik/tikhup/service/TikHupServiceImpl.java |   37 +-
 .../tik/voice/client/CosyVoiceClient.java     |  178 ++
 .../tik/voice/client/LatentsyncClient.java    |  141 ++
 .../voice/client/dto/CosyVoiceTtsRequest.java |   54 +
 .../voice/client/dto/CosyVoiceTtsResult.java  |   37 +
 .../client/dto/LatentsyncSubmitRequest.java   |   34 +
 .../client/dto/LatentsyncSubmitResponse.java  |   39 +
 .../tik/voice/config/CosyVoiceProperties.java |   74 +
 .../voice/config/LatentsyncProperties.java    |   78 +
 .../AppTikLatentsyncController.java           |   38 +
 .../controller/AppTikUserVoiceController.java |   95 +
 .../voice/dal/dataobject/TikUserVoiceDO.java  |   59 +
 .../voice/dal/mysql/TikUserVoiceMapper.java   |   26 +
 .../tik/voice/service/LatentsyncService.java  |   20 +
 .../voice/service/LatentsyncServiceImpl.java  |   42 +
 .../voice/service/TikUserVoiceService.java    |   75 +
 .../service/TikUserVoiceServiceImpl.java      |  864 +++++++++
 .../voice/util/ByteArrayMultipartFile.java    |   69 +
 .../voice/vo/AppTikLatentsyncSubmitReqVO.java |   37 +
 .../vo/AppTikLatentsyncSubmitRespVO.java      |   22 +
 .../voice/vo/AppTikUserVoiceCreateReqVO.java  |   38 +
 .../voice/vo/AppTikUserVoicePageReqVO.java    |   23 +
 .../tik/voice/vo/AppTikUserVoiceRespVO.java   |   48 +
 .../voice/vo/AppTikUserVoiceUpdateReqVO.java  |   36 +
 .../tik/voice/vo/AppTikVoicePreviewReqVO.java |   43 +
 .../voice/vo/AppTikVoicePreviewRespVO.java    |   26 +
 .../tik/voice/vo/AppTikVoiceTtsReqVO.java     |   46 +
 .../tik/voice/vo/AppTikVoiceTtsRespVO.java    |   29 +
 .../service/LatentsyncServiceImplTest.java    |   62 +
 .../src/main/resources/application-local.yaml |    7 +-
 .../src/main/resources/application.yaml       |    7 +
 52 files changed, 4627 insertions(+), 2251 deletions(-)
 create mode 100644 CLAUDE.md
 create mode 100644 frontend/app/web-gold/src/api/mix.js
 create mode 100644 frontend/app/web-gold/src/api/voice.js
 delete mode 100644 frontend/app/web-gold/src/views/mix/MixEditor.vue
 delete mode 100644 yudao-module-tik/DESIGN.md
 delete mode 100644 yudao-module-tik/LOGIC_ANALYSIS.md
 delete mode 100644 yudao-module-tik/LOGIC_REVIEW.md
 delete mode 100644 yudao-module-tik/UPLOAD_STRATEGY.md
 create mode 100644 yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/client/CosyVoiceClient.java
 create mode 100644 yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/client/LatentsyncClient.java
 create mode 100644 yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/client/dto/CosyVoiceTtsRequest.java
 create mode 100644 yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/client/dto/CosyVoiceTtsResult.java
 create mode 100644 yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/client/dto/LatentsyncSubmitRequest.java
 create mode 100644 yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/client/dto/LatentsyncSubmitResponse.java
 create mode 100644 yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/config/CosyVoiceProperties.java
 create mode 100644 yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/config/LatentsyncProperties.java
 create mode 100644 yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/controller/AppTikLatentsyncController.java
 create mode 100644 yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/controller/AppTikUserVoiceController.java
 create mode 100644 yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/dal/dataobject/TikUserVoiceDO.java
 create mode 100644 yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/dal/mysql/TikUserVoiceMapper.java
 create mode 100644 yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/service/LatentsyncService.java
 create mode 100644 yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/service/LatentsyncServiceImpl.java
 create mode 100644 yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/service/TikUserVoiceService.java
 create mode 100644 yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/service/TikUserVoiceServiceImpl.java
 create mode 100644 yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/util/ByteArrayMultipartFile.java
 create mode 100644 yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikLatentsyncSubmitReqVO.java
 create mode 100644 yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikLatentsyncSubmitRespVO.java
 create mode 100644 yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikUserVoiceCreateReqVO.java
 create mode 100644 yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikUserVoicePageReqVO.java
 create mode 100644 yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikUserVoiceRespVO.java
 create mode 100644 yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikUserVoiceUpdateReqVO.java
 create mode 100644 yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikVoicePreviewReqVO.java
 create mode 100644 yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikVoicePreviewRespVO.java
 create mode 100644 yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikVoiceTtsReqVO.java
 create mode 100644 yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikVoiceTtsRespVO.java
 create mode 100644 yudao-module-tik/src/test/java/cn/iocoder/yudao/module/tik/voice/service/LatentsyncServiceImplTest.java

diff --git a/.cursor/rules/backend.mdc b/.cursor/rules/backend.mdc
index 102be8e118..5aba776dd9 100644
--- a/.cursor/rules/backend.mdc
+++ b/.cursor/rules/backend.mdc
@@ -88,24 +88,7 @@ yudao-module-{模块名}/
 
 ### 目录结构示例
 
-#### 示例 1：简单模块（tikhup）
-```
-tikhup/
-├── controller/
-│   └── TikHupController.java
-├── service/
-│   ├── TikHupService.java
-│   ├── TikHupServiceImpl.java
-│   └── TikFileTransCharacters.java
-├── mapper/
-│   ├── TikPromptMapper.java
-│   └── TikTokenMapper.java
-└── vo/
-    ├── TikPromptVO.java
-    └── TikTokenVO.java
-```
-
-#### 示例 2：完整模块（file）
+#### 示例 2：模块
 ```
 file/
 ├── controller/
@@ -131,8 +114,8 @@ file/
 ### 目录结构原则
 
 1. **统一性**：同一模块内保持结构一致
-2. **简洁性**：使用 `mapper/` 和 `dataobject/` 包，结构清晰
-3. **可选性**：没有 DO 对象时可以省略 `dataobject/` 包
+2. **简洁性**：使用 `mapper/`，结构清晰
+3. **可选性**：省略 `dataobject/` 包
 4. **可扩展性**：预留扩展空间，便于后续功能扩展
 
 ## Controller 层规范
diff --git a/CLAUDE.md b/CLAUDE.md
new file mode 100644
index 0000000000..d27ddc69a1
--- /dev/null
+++ b/CLAUDE.md
@@ -0,0 +1,511 @@
+# CLAUDE.md
+
+本文档为 Claude Code (claude.ai/code) 在此仓库中处理代码提供指导。
+
+## 项目概览
+
+**Yudao（芋道）** - 基于 Spring Boot 的快速开发平台，采用多模块架构。这是 Yudao 平台的 AI/媒体重点部署版本，具备数字人生成、语音克隆、视频混剪和内容分析能力。
+
+### 核心技术栈
+
+**后端：**
+- Java 17 + Spring Boot 3.5.5
+- Maven 构建管理
+- MyBatis Plus 3.5.14 + Dynamic Datasource ORM
+- Redis + Redisson 缓存
+- Spring Security 6.5.2 认证
+- Flowable 7.0.1 工作流
+- Springdoc/OpenAPI 文档
+
+**前端：**
+- Vue.js 3.5.22 + Composition API
+- Vite 7.1.7 构建工具
+- Ant Design Vue 4.2.6 UI组件
+- TypeScript 类型安全
+- Pinia 3.0.3 状态管理
+- TailwindCSS 4.1.14 样式
+
+**数据库与基础设施：**
+- MySQL 8.0+（主要）
+- 支持 PostgreSQL、Oracle、SQL Server、DM、KingbaseES、OpenGauss、TiDB
+- Redis 缓存
+- Docker 容器化
+
+## 项目结构
+
+```
+/d/projects/sionrui/
+├── yudao-dependencies/          # Maven 依赖版本管理
+├── yudao-framework/             # 框架组件和 Spring Boot 启动器
+├── yudao-server/                # 主应用服务器（端口 9900）
+├── yudao-module-system/         # 系统管理（用户、角色、权限）
+├── yudao-module-infra/          # 基础设施（文件、配置、任务）
+├── yudao-module-member/         # 会员中心
+├── yudao-module-pay/            # 支付系统
+├── yudao-module-ai/             # AI/ML 功能（聊天、图像、知识、音乐）
+├── yudao-module-tik/            # Tik/媒体模块（语音克隆、头像、视频）
+├── frontend/app/web-gold/       # Vue.js 前端
+├── sql/                         # 数据库模式
+├── script/                      # 构建和部署脚本
+└── docs/                        # 文档
+```
+
+## 常用开发命令
+
+### 后端（Maven）
+
+**构建和运行：**
+```bash
+# 构建项目
+mvn clean package -DskipTests
+
+# 运行特定模块的测试
+mvn test -pl yudao-module-tik
+
+# 启动服务器
+cd yudao-server && mvn spring-boot:run -Dspring-boot.run.profiles=local
+
+# 使用特定配置构建
+mvn clean package -Pdev -DskipTests
+```
+
+**代码生成：**
+- 内置 CRUD 操作代码生成器
+- 生成 Java、Vue、SQL 脚本和 API 文档
+- 支持单表、树表、主子表模式
+
+### 前端（Vue.js）
+
+**开发：**
+```bash
+cd frontend/app/web-gold
+
+# 安装依赖
+npm install
+
+# 启动开发服务器（代理到后端 9900 端口）
+npm run dev
+
+# 生产构建
+npm run build
+
+# 代码检查
+npm run lint
+
+# 代码格式化
+npm run format
+```
+
+**可用脚本：**
+- `dev` - 带热重载的开发服务器
+- `build` - 生产构建
+- `preview` - 预览生产构建
+- `lint:oxlint` - 运行 OxLint 并自动修复
+- `lint:eslint` - 运行 ESLint 并自动修复
+- `lint` - 运行所有检查器
+- `format` - 使用 Prettier 格式化代码
+
+### Docker
+
+**使用 Docker Compose：**
+```bash
+# 启动所有服务（MySQL、Redis、Server、Admin）
+cd script/docker
+docker-compose up -d
+
+# 启动特定服务
+docker-compose up -d mysql redis
+```
+
+**手动 Docker 构建：**
+```bash
+# 后端
+cd yudao-server
+docker build -t yudao-server .
+
+# 前端
+cd frontend/app/web-gold
+docker build -t web-gold .
+```
+
+## 模块架构
+
+### 后端模块结构模式
+
+每个模块都遵循一致的分层架构：
+```
+module/
+├── controller/      # REST 控制器（admin-api/、app/）
+├── service/         # 业务逻辑 + 接口
+│   ├── {Xxx}Service.java           # 接口
+│   └── {Xxx}ServiceImpl.java       # 实现
+├── dal/             # 数据访问层
+│   ├── mysql/       # MyBatis Mappers 和 DO 类
+│   └── redis/       # Redis 操作
+├── client/          # 外部 API 客户端
+├── config/          # 配置类
+├── util/            # 工具类
+└── vo/              # 值对象
+    ├── {Xxx}SaveReqVO.java         # 创建请求
+    ├── {Xxx}PageReqVO.java         # 分页请求
+    ├── {Xxx}UpdateReqVO.java       # 更新请求
+    └── {Xxx}RespVO.java            # 响应
+```
+
+**核心模块：**
+
+1. **yudao-module-tik** - 媒体/AI 功能
+   - `voice/` - 语音克隆（CosyVoice、Latentsync）
+   - `file/` - 带 OSS 集成的文件管理
+   - `chat/` - 对话管理
+   - `media/` - 媒体处理
+   - `quota/` - 配额管理
+
+2. **yudao-module-ai** - AI/ML 能力
+   - 聊天补全 API
+   - 图像生成（Midjourney）
+   - 音乐生成（Suno）
+   - 带向量搜索的知识库
+
+3. **yudao-module-system** - 核心系统功能
+   - 用户/角色/权限管理
+   - 多租户支持
+   - 审计日志
+
+### 前端结构
+
+```
+frontend/app/web-gold/src/
+├── api/              # API 服务层
+│   ├── axios/        # Axios 拦截器
+│   ├── voice.js      # 语音相关 API
+│   └── mix.js        # 视频混剪 API
+├── components/       # 可复用 Vue 组件
+├── router/
+│   └── index.js      # Vue Router 配置
+├── stores/
+│   └── voiceCopy.js  # Pinia 状态管理
+├── views/
+│   ├── dh/           # 数字人功能
+│   │   ├── Avatar.vue
+│   │   ├── Video.vue
+│   │   └── VoiceCopy.vue
+│   ├── material/     # 素材库
+│   └── content-style/# 内容分析
+└── utils/
+    └── video-cover.ts # 工具函数
+```
+
+**核心路由：**
+- `/digital-human/*` - 语音克隆、头像、视频生成
+- `/content-style/*` - 内容分析和基准测试
+- `/trends/*` - 趋势分析
+- `/material/*` - 素材库管理
+
+## 配置
+
+### 后端配置文件
+
+**主配置：** `yudao-server/src/main/resources/application.yaml`
+- Spring Boot 配置
+- 数据库连接
+- Redis 设置
+- 安全设置
+- 多租户配置
+- AI 服务 API 密钥
+
+**本地开发：** `yudao-server/src/main/resources/application-local.yaml`
+- 本地开发覆盖
+- 数据库：`jdbc:mysql://8.155.172.147:3306/sion_rui_dev`
+- Redis：`8.155.172.147:6379`
+- 端口：9900
+
+**配置环境：**
+- `local` - 开发（端口 9900）
+- `dev` - 开发服务器
+- `prod` - 生产
+
+### 前端配置
+
+**Vite 配置：** `frontend/app/web-gold/vite.config.js`
+- 开发服务器代理到后端
+- 构建配置
+- 插件设置
+
+**API 代理：**
+- 开发服务器将 `/admin-api` 和 `/api` 代理到 `http://localhost:9900`
+
+## 数据库模式
+
+**位置：** `sql/mysql/`
+- 主模式：`ruoyi-vue-pro.sql` (949KB)
+- Quartz：`quartz.sql` 用于定时任务
+- 模块特定迁移在各模块文件夹中
+
+**模式更新：**
+- 将 SQL 迁移添加到 `sql/mysql/`
+- 遵循命名约定：`V{version}__{description}.sql`
+
+## API 文档
+
+- **Swagger UI：** `http://localhost:9900/swagger-ui.html`
+- **API 文档：** `http://localhost:9900/v3/api-docs`
+
+**API 路径约定：**
+- 管理 API：`/admin-api/{module}/{resource}`
+- 应用 API：`/api/{module}/{resource}`
+- CRUD 端点：
+  - 创建：`POST /module/resource/create`
+  - 更新：`PUT /module/resource/update`
+  - 删除：`DELETE /module/resource/delete`
+  - 查询：`GET /module/resource/get?id=xxx`
+  - 分页：`GET /module/resource/page`
+
+## 代码风格与规范
+
+### 后端（Java）
+
+**架构层：**
+1. **Controller** - 请求处理、验证、调用 Service
+2. **Service** - 业务逻辑、事务管理
+3. **Mapper** - 使用 MyBatis Plus 进行数据访问
+4. **VO** - API 请求/响应对象
+5. **DO** - 映射到数据库表的数据对象
+
+**关键规范：**
+- Mapper 接口继承 `BaseMapperX<T>`
+- DO 类继承 `BaseDO` 或 `TenantBaseDO` 以支持多租户
+- 使用 `@PreAuthorize` 进行权限控制
+- 统一使用 `CommonResult<T>` 作为 API 响应
+- Service 方法使用 `@Transactional` 进行写操作
+- 异常代码在 `ErrorCodeConstants` 中，格式为：`MODULE_RESOURCE_ACTION_ERROR`
+
+**命名规范：**
+- Controller：`{Xxx}Controller` 或 `App{Xxx}Controller`
+- Service：`{Xxx}Service` 和 `{Xxx}ServiceImpl`
+- Mapper：`{Xxx}Mapper`
+- VO：`{Xxx}SaveReqVO`、`{Xxx}PageReqVO`、`{Xxx}RespVO`
+- DO：`{Xxx}DO`
+
+### 前端（Vue.js）
+
+**关键模式：**
+- Composition API + `<script setup>`
+- Pinia 进行状态管理和持久化
+- Axios 拦截器处理认证和租户
+- TypeScript 提供类型安全
+
+**代码检查：**
+- ESLint + OxLint 保证代码质量
+- Prettier 进行代码格式化
+- 提交前运行 `npm run lint`
+
+## 测试
+
+**后端：**
+- JUnit 5 + Mockito 进行单元测试
+- Spring Boot Test 进行集成测试
+- 测试位置：`src/test/java`
+- 运行测试：`mvn test`
+
+**前端：**
+- Vitest 进行单元测试
+- Cypress 进行端到端测试
+- 运行测试：`npm run test`
+
+## 部署
+
+**生产部署：**
+```bash
+# 使用部署脚本
+cd script/shell
+./deploy.sh
+
+# 手动部署
+# 1. 构建 JAR
+mvn clean package -DskipTests -Pprod
+
+# 2. 部署到服务器
+# deploy.sh 脚本处理：
+# - 备份前一版本
+# - 停止当前服务
+# - 传输新 JAR
+# - 启动服务
+# - 健康检查
+```
+
+**JVM 选项：**
+- 默认：`-Xms512m -Xmx512m -XX:+HeapDumpOnOutOfMemoryError`
+- 可在 `deploy.sh` 中配置
+
+**健康检查：**
+- 端点：`/actuator/health`
+- 端口：48080（生产）或 9900（本地）
+
+## 开发工作流
+
+### 创建新模块
+
+1. 按照标准目录结构创建模块
+2. 将 SQL 迁移添加到 `sql/mysql/`
+3. 运行代码生成器进行 CRUD 操作
+4. 实现 Controller、Service、Mapper 层
+5. 编写单元测试
+6. 更新 API 文档
+
+### 添加新 API 端点
+
+1. 在模块的 `vo/` 包中创建 VO 类
+2. 使用适当注解实现 Controller：
+   - `@Tag`、`@Operation` 用于 Swagger 文档
+   - `@PreAuthorize` 用于权限
+   - `@Valid` 用于验证
+3. 实现包含业务逻辑的 Service 层
+4. 如需要，创建/更新 Mapper
+5. 通过 Swagger UI 测试端点
+
+### 前端开发
+
+1. 在 `src/api/` 中创建/更新 API 服务
+2. 在 `src/stores/` 中添加 Pinia 存储（如需要）
+3. 在 `src/views/` 中创建 Vue 组件
+4. 更新 `src/router/index.js` 中的路由
+5. 使用 `npm run dev` 测试
+
+## 多租户
+
+- 配置中默认启用
+- DO 类继承 `TenantBaseDO` 实现租户隔离
+- 框架自动注入 `tenantId`
+- 需要时使用 `@TenantIgnore` 覆盖
+
+**配置：**
+```yaml
+yudao:
+  tenant:
+    enable: true
+    ignore-urls:
+      - /jmreport/*
+    ignore-tables:
+      - table_name
+```
+
+## 安全
+
+**认证与授权：**
+- Spring Security 认证框架
+- 基于令牌的身份认证
+- 基于角色的访问控制（RBAC）
+- 权限格式：`module:resource:action`
+
+**API 安全：**
+- 支持 API 加密（AES/RSA）
+- 配置中的请求/响应加密密钥
+- 可配置的 XSS 保护
+
+**数据保护：**
+- 字段级加密支持
+- 通过 MyBatis Plus 防止 SQL 注入
+- 使用 `@Valid` 进行参数验证
+
+## 缓存
+
+**Redis 配置：**
+- 缓存类型：`REDIS`
+- 默认 TTL：1 小时
+- 连接：`8.155.172.147:6379`
+
+**缓存模式：**
+- 键格式：`模块:资源:id`
+- 使用 `@Cacheable`、`@CacheEvict` 管理缓存
+- 热数据缓存提升性能
+
+## 集成点
+
+**AI 服务：**
+- CosyVoice：语音克隆
+- Latentsync：语音合成
+- Midjourney：图像生成
+- Suno：音乐生成
+- 多个 LLM 提供商（OpenAI、Claude、Gemini 等）
+
+**文件存储：**
+- S3 兼容（MinIO、AWS S3 等）
+- 本地存储
+- FTP
+- 数据库存储
+
+**消息队列：**
+- Redis（Pub/Sub、Stream）
+- Kafka
+- RabbitMQ
+- RocketMQ
+
+## 监控与可观测性
+
+**Actuator 端点：**
+- `/actuator/health` - 健康检查
+- `/actuator/metrics` - 指标
+- `/actuator/env` - 环境属性
+
+**监控工具：**
+- Spring Boot Admin 应用程序监控
+- SkyWalking 分布式追踪
+- Druid SQL 监控
+
+## Cursor 规则集成
+
+此仓库在 `.cursor/rules/backend.mdc` 中配置了 **Cursor 规则** 用于 Spring Boot 开发最佳实践。主要规则包括：
+
+- 分层架构强制执行（Controller → Service → Mapper）
+- 模块结构约定
+- VO/DO 命名标准
+- 事务管理模式
+- 多租户支持
+- API 路径约定
+- 安全和权限模式
+
+## 重要注意事项
+
+1. **数据库：** 需要 MySQL 8.0+，外部连接配置在 `application-local.yaml`
+2. **Redis：** 缓存和会话管理所需
+3. **Java 版本：** 需要 JDK 17+
+4. **Node 版本：** 前端需要 Node.js 20.19.0+ 或 22.12.0+
+5. **端口：** 后端默认 9900，前端默认 5173
+6. **API 密钥：** `application.yaml` 中配置了多个 AI 服务 API 密钥 - 不要提交到公共仓库
+7. **多租户：** 默认启用 - 所有 DO 类应继承 `TenantBaseDO`
+
+## 故障排除
+
+**常见问题：**
+
+1. **端口已被占用：**
+   - 在 `application-local.yaml` 中更改端口：`server.port=9999`
+   - 或终止进程：`lsof -ti:9900 | xargs kill -9`
+
+2. **数据库连接失败：**
+   - 验证 MySQL 运行：`mysql -h 8.155.172.147 -u root -p`
+   - 检查 `application-local.yaml` 中的连接设置
+
+3. **Redis 连接失败：**
+   - 验证 Redis：`redis-cli -h 8.155.172.147 -p 6379`
+   - 检查密码/认证设置
+
+4. **前端构建错误：**
+   - 清理 node_modules：`rm -rf node_modules package-lock.json`
+   - 重新安装：`npm install`
+   - 检查 Node 版本：`node --version`
+
+5. **Maven 构建错误：**
+   - 清理构建：`mvn clean install`
+   - 跳过测试：`mvn clean package -DskipTests`
+
+## 资源
+
+- **官方文档：** https://doc.iocoder.cn/
+- **快速开始：** https://doc.iocoder.cn/quick-start/
+- **视频教程：** https://doc.iocoder.cn/video/
+- **API 文档：** http://localhost:9900/swagger-ui.html（运行时）
+- **Spring Boot 参考：** https://docs.spring.io/spring-boot/docs/current/reference/html/
+- **Vue.js 指南：** https://vuejs.org/guide/
+- **Yudao GitHub：** https://github.com/YunaiV/ruoyi-vue-pro
diff --git a/frontend/app/web-gold/src/api/mix.js b/frontend/app/web-gold/src/api/mix.js
new file mode 100644
index 0000000000..49c775c1b0
--- /dev/null
+++ b/frontend/app/web-gold/src/api/mix.js
@@ -0,0 +1,46 @@
+/**
+ * 混剪 API 服务
+ */
+
+import http from './http'
+import { API_BASE } from '@gold/config/api'
+
+const BASE_URL = `${API_BASE.APP}/api/media`
+
+/**
+ * 提交素材混剪任务
+ * @param {Object} data
+ * @param {string} data.title
+ * @param {string} data.text
+ * @param {string[]} data.videoUrls
+ * @param {string[]} data.bgMusicUrls
+ * @param {number} data.produceCount
+ */
+export const MixService = {
+  batchProduceAlignment({ title, text, videoUrls = [], bgMusicUrls = [], produceCount = 1 }) {
+    const formData = new URLSearchParams()
+    formData.append('title', title)
+    formData.append('text', text)
+    videoUrls.forEach((url) => {
+      if (url) {
+        formData.append('videoArray', url)
+      }
+    })
+    bgMusicUrls.forEach((url) => {
+      if (url) {
+        formData.append('bgMusicArray', url)
+      }
+    })
+    formData.append('produceCount', produceCount)
+
+    return http.post(`${BASE_URL}/batchProduceAlignment`, formData, {
+      headers: {
+        'Content-Type': 'application/x-www-form-urlencoded'
+      }
+    })
+  }
+}
+
+export default MixService
+
+
diff --git a/frontend/app/web-gold/src/api/voice.js b/frontend/app/web-gold/src/api/voice.js
new file mode 100644
index 0000000000..8c8908cbaf
--- /dev/null
+++ b/frontend/app/web-gold/src/api/voice.js
@@ -0,0 +1,110 @@
+/**
+ * 配音 API 服务
+ * 对应后端 tik 模块的配音管理接口
+ */
+
+import http from './http'
+import { API_BASE } from '@gold/config/api'
+
+// 统一使用 /api/tik 前缀
+const BASE_URL = `${API_BASE.APP}/api/tik/voice`
+
+/**
+ * 配音 API 服务
+ */
+export const VoiceService = {
+  /**
+   * 创建配音
+   * @param {Object} data - 请求数据
+   * @param {string} data.name - 配音名称（必填）
+   * @param {number} data.fileId - 音频文件编号（必填）
+   * @param {boolean} data.autoTranscribe - 是否自动识别（可选）
+   * @param {string} data.language - 语言（可选）
+   * @param {string} data.gender - 音色类型（可选）
+   * @param {string} data.note - 备注（可选）
+   * @returns {Promise}
+   */
+  create(data) {
+    return http.post(`${BASE_URL}/create`, data)
+  },
+
+  /**
+   * 更新配音
+   * @param {Object} data - 请求数据
+   * @param {number} data.id - 配音编号（必填）
+   * @param {string} data.name - 配音名称（可选）
+   * @param {string} data.language - 语言（可选）
+   * @param {string} data.gender - 音色类型（可选）
+   * @param {string} data.note - 备注（可选）
+   * @returns {Promise}
+   */
+  update(data) {
+    return http.put(`${BASE_URL}/update`, data)
+  },
+
+  /**
+   * 删除配音
+   * @param {number} id - 配音编号
+   * @returns {Promise}
+   */
+  delete(id) {
+    return http.delete(`${BASE_URL}/delete`, {
+      params: { id }
+    })
+  },
+
+  /**
+   * 分页查询配音列表
+   * @param {Object} params - 查询参数
+   * @param {number} params.pageNo - 页码
+   * @param {number} params.pageSize - 每页数量
+   * @param {string} params.name - 配音名称（模糊查询）
+   * @returns {Promise}
+   */
+  getPage(params) {
+    return http.get(`${BASE_URL}/page`, { params })
+  },
+
+  /**
+   * 获取单个配音
+   * @param {number} id - 配音编号
+   * @returns {Promise}
+   */
+  get(id) {
+    return http.get(`${BASE_URL}/get`, {
+      params: { id }
+    })
+  },
+
+  /**
+   * 手动触发语音识别
+   * @param {number} id - 配音编号
+   * @returns {Promise}
+   */
+  transcribe(id) {
+    return http.post(`${BASE_URL}/transcribe`, null, {
+      params: { id }
+    })
+  },
+
+  /**
+   * 文本转语音（CosyVoice）
+   * @param {Object} data
+   * @returns {Promise}
+   */
+  synthesize(data) {
+    return http.post(`${BASE_URL}/tts`, data)
+  },
+
+  /**
+   * 我的音色试听
+   * @param {Object} data
+   * @returns {Promise}
+   */
+  preview(data) {
+    return http.post(`${BASE_URL}/preview`, data)
+  }
+}
+
+export default VoiceService
+
diff --git a/frontend/app/web-gold/src/components/ChatMessageRenderer.vue b/frontend/app/web-gold/src/components/ChatMessageRenderer.vue
index 7fc24cd261..57afcf98ad 100644
--- a/frontend/app/web-gold/src/components/ChatMessageRenderer.vue
+++ b/frontend/app/web-gold/src/components/ChatMessageRenderer.vue
@@ -3,7 +3,7 @@
 </template>
 
 <script setup>
-import { ref, watch, onUnmounted } from 'vue'
+import { ref, watch } from 'vue'
 import { renderMarkdown } from '@/utils/markdown'
 
 const props = defineProps({
@@ -17,152 +17,73 @@ const props = defineProps({
   }
 })
 
-// 当前显示的内容（用于打字机效果）
-const displayedContent = ref('')
-// 目标内容（完整内容）
-const targetContent = ref('')
-// 动画帧 ID
-let animationFrameId = null
-// 打字机速度（字符/帧，可根据性能调整）
-const TYPING_SPEED = 3
-// 是否正在执行打字机动画
-const isTyping = ref(false)
-
-/**
- * 高性能打字机效果渲染
- * 使用 requestAnimationFrame 实现平滑的逐字符显示
- */
-function typewriterEffect() {
-  if (!isTyping.value) return
-  
-  const currentLength = displayedContent.value.length
-  const targetLength = targetContent.value.length
-  
-  if (currentLength >= targetLength) {
-    // 已完成，停止动画
-    isTyping.value = false
-    displayedContent.value = targetContent.value
-    return
-  }
-  
-  // 每次增加多个字符以提高性能
-  const nextLength = Math.min(currentLength + TYPING_SPEED, targetLength)
-  displayedContent.value = targetContent.value.slice(0, nextLength)
-  
-  // 继续下一帧
-  animationFrameId = requestAnimationFrame(typewriterEffect)
-}
-
-/**
- * 开始打字机效果
- */
-function startTypewriter() {
-  if (animationFrameId) {
-    cancelAnimationFrame(animationFrameId)
-  }
-  
-  isTyping.value = true
-  animationFrameId = requestAnimationFrame(typewriterEffect)
-}
-
-/**
- * 停止打字机效果
- */
-function stopTypewriter() {
-  if (animationFrameId) {
-    cancelAnimationFrame(animationFrameId)
-    animationFrameId = null
-  }
-  isTyping.value = false
-}
-
-/**
- * 计算渲染的 HTML 内容
- */
+// 当前渲染的内容（避免重复渲染）
+const currentContent = ref('')
+// 渲染的 HTML 内容
 const renderedContent = ref('')
 
 /**
  * 更新渲染内容
+ * 只有当内容真正改变时才更新，避免重复渲染
  */
 function updateRenderedContent() {
-  const content = displayedContent.value
+  const content = currentContent.value
+
+  // 避免重复渲染相同内容
+  if (content === renderedContent.value.replace(/<[^>]*>/g, '')) {
+    return
+  }
+
   if (!content) {
     renderedContent.value = ''
     return
   }
-  
+
   // 渲染 markdown 为 HTML
   renderedContent.value = renderMarkdown(content)
 }
 
-// 监听 displayedContent 变化，更新渲染
-watch(displayedContent, () => {
-  updateRenderedContent()
-}, { immediate: true })
-
 /**
  * 处理内容更新
+ * 普通流式渲染：直接显示所有内容，不使用打字机效果
  */
 function handleContentUpdate(newContent) {
   if (!newContent) {
-    targetContent.value = ''
-    displayedContent.value = ''
-    stopTypewriter()
+    currentContent.value = ''
+    updateRenderedContent()
     return
   }
-  
-  // 更新目标内容
-  targetContent.value = newContent
-  
-  if (props.isStreaming) {
-    // 流式传输模式：使用打字机效果显示内容
-    // 流式传输时，内容会逐步到达，使用打字机效果增强体验
-    const currentLength = displayedContent.value.length
-    const newLength = newContent.length
-    
-    if (newLength !== currentLength) {
-      // 内容发生变化
-      if (newLength < currentLength) {
-        // 内容被重置或缩短，直接显示新内容
-        displayedContent.value = newContent
-        stopTypewriter()
-      } else {
-        // 内容增加，使用打字机效果显示新增部分
-        if (!isTyping.value) {
-          startTypewriter()
-        }
-      }
-    }
-  } else {
-    // 静态内容模式：直接显示全部内容，不使用打字机效果
-    displayedContent.value = newContent
-    stopTypewriter()
-  }
+
+  // 更新当前内容
+  currentContent.value = newContent
+  updateRenderedContent()
 }
 
-// 监听 content 变化
+// 监听 content 变化，使用防抖处理避免频繁更新
+let updateTimeout = null
 watch(() => props.content, (newContent) => {
-  handleContentUpdate(newContent)
-}, { immediate: true })
+  // 清除之前的定时器
+  if (updateTimeout) {
+    clearTimeout(updateTimeout)
+  }
+
+  // 延迟更新，避免流式传输时频繁更新导致的性能问题
+  updateTimeout = setTimeout(() => {
+    handleContentUpdate(newContent)
+  }, 50) // 50ms 防抖
+})
 
 // 监听 isStreaming 变化
 watch(() => props.isStreaming, (newVal, oldVal) => {
-  if (!newVal && oldVal) {
-    // 流式传输结束，确保显示完整内容，停止打字机效果
-    if (targetContent.value) {
-      displayedContent.value = targetContent.value
-    }
-    stopTypewriter()
-  } else if (newVal) {
-    // 开始流式传输，如果内容有变化，启动打字机效果
-    // 打字机效果会在 handleContentUpdate 中根据内容变化自动启动
+  // 流式传输结束时，确保显示完整内容
+  if (!newVal && oldVal && props.content) {
+    currentContent.value = props.content
+    updateRenderedContent()
   }
 })
 
-// 组件卸载时清理
-onUnmounted(() => {
-  stopTypewriter()
-})
+// 立即渲染初始内容
+handleContentUpdate(props.content)
 </script>
 
 <style scoped>
diff --git a/frontend/app/web-gold/src/router/index.js b/frontend/app/web-gold/src/router/index.js
index 95bae9a27a..cccef4b493 100644
--- a/frontend/app/web-gold/src/router/index.js
+++ b/frontend/app/web-gold/src/router/index.js
@@ -55,7 +55,6 @@ const routes = [
     ]
   },
   { path: '/realtime-hot', name: '实时热点推送', component: () => import('../views/realtime/RealtimeHot.vue') },
-  { path: '/mix-editor', name: '素材混剪', component: () => import('../views/mix/MixEditor.vue') },
   { path: '/capcut-import', name: '剪映导入', component: () => import('../views/capcut/CapcutImport.vue') },
   { path: '/help', name: '帮助', component: () => import('../views/misc/Help.vue') },
   { path: '/download', name: '下载', component: () => import('../views/misc/Download.vue') },
diff --git a/frontend/app/web-gold/src/stores/voiceCopy.js b/frontend/app/web-gold/src/stores/voiceCopy.js
index 1a7144c082..9c66c24855 100644
--- a/frontend/app/web-gold/src/stores/voiceCopy.js
+++ b/frontend/app/web-gold/src/stores/voiceCopy.js
@@ -1,13 +1,13 @@
 import { defineStore } from 'pinia'
-import storage from '@/utils/storage'
-
-const STORAGE_KEY = 'cosy_voice_profiles'
+import { VoiceService } from '@/api/voice'
+import { message } from 'ant-design-vue'
 
 export const useVoiceCopyStore = defineStore('voiceCopy', {
   state: () => ({
     profiles: [],
     activeId: '',
     loaded: false,
+    loading: false
   }),
   getters: {
     activeProfile(state) {
@@ -15,48 +15,158 @@ export const useVoiceCopyStore = defineStore('voiceCopy', {
     }
   },
   actions: {
-    generateId() {
-      return `${Date.now()}_${Math.floor(Math.random() * 1e6)}`
-    },
+    /**
+     * 加载配音列表
+     */
     async load() {
-      if (this.loaded) return
-      const list = await storage.getJSON(STORAGE_KEY, [])
-      this.profiles = Array.isArray(list) ? list : []
-      if (!this.activeId && this.profiles.length) this.activeId = this.profiles[0].id
-      this.loaded = true
-    },
-    async persist() {
-      await storage.setJSON(STORAGE_KEY, this.profiles)
+      if (this.loaded && !this.loading) return
+      
+      this.loading = true
+      try {
+        const res = await VoiceService.getPage({
+          pageNo: 1,
+          pageSize: 100 // 加载所有数据
+        })
+        
+        if (res.code === 0) {
+          this.profiles = (res.data.list || []).map((item) => ({
+            ...item,
+            voiceId: item.voiceId || '',
+            transcription: item.transcription || '',
+            fileUrl: item.fileUrl || ''
+          }))
+          // 如果有数据且没有选中项，选中第一个
+          if (!this.activeId && this.profiles.length > 0) {
+            this.activeId = this.profiles[0].id
+          }
+          this.loaded = true
+        } else {
+          message.error(res.msg || '加载失败')
+        }
+      } catch (error) {
+        console.error('加载配音列表失败:', error)
+        message.error('加载失败，请稍后重试')
+      } finally {
+        this.loading = false
+      }
     },
+
+    /**
+     * 添加配音
+     */
     async add(profile) {
-      const id = this.generateId()
-      const name = profile.name || `克隆语音-${this.profiles.length + 1}`
-      const payload = { ...profile, id, name }
-      this.profiles.unshift(payload)
-      this.activeId = id
-      await this.persist()
-      return payload
+      try {
+        const res = await VoiceService.create({
+          name: profile.name || `克隆语音-${this.profiles.length + 1}`,
+          fileId: profile.fileId,
+          autoTranscribe: profile.autoTranscribe || false,
+          language: profile.language || 'zh-CN',
+          gender: profile.gender || 'female',
+          note: profile.note || ''
+        })
+        
+        if (res.code === 0) {
+          const newProfile = {
+            ...profile,
+            id: res.data
+          }
+          this.profiles.unshift(newProfile)
+          this.activeId = newProfile.id
+          await this.load() // 重新加载以获取完整数据
+          return newProfile
+        } else {
+          message.error(res.msg || '创建失败')
+          throw new Error(res.msg || '创建失败')
+        }
+      } catch (error) {
+        console.error('添加配音失败:', error)
+        throw error
+      }
     },
+
+    /**
+     * 更新配音
+     */
     async update(profile) {
-      const idx = this.profiles.findIndex(p => p.id === profile.id)
-      if (idx === -1) return await this.add({ ...profile, id: '' })
-      this.profiles[idx] = { ...profile }
-      await this.persist()
-      return this.profiles[idx]
+      if (!profile.id) {
+        return await this.add(profile)
+      }
+      
+      try {
+        const res = await VoiceService.update({
+          id: profile.id,
+          name: profile.name,
+          language: profile.language,
+          gender: profile.gender,
+          note: profile.note
+        })
+        
+        if (res.code === 0) {
+          const index = this.profiles.findIndex(p => p.id === profile.id)
+          if (index > -1) {
+            // 重新加载以获取最新数据
+            await this.load()
+            const updated = this.profiles.find(p => p.id === profile.id)
+            return updated || profile
+          }
+          return profile
+        } else {
+          message.error(res.msg || '更新失败')
+          throw new Error(res.msg || '更新失败')
+        }
+      } catch (error) {
+        console.error('更新配音失败:', error)
+        throw error
+      }
     },
+
+    /**
+     * 复制配音
+     */
     async duplicate(profile, name) {
-      const copy = { ...profile, id: '', name }
+      const copy = {
+        ...profile,
+        id: null,
+        name: name || `${profile.name}-副本`
+      }
       return await this.add(copy)
     },
+
+    /**
+     * 删除配音
+     */
     async remove(id) {
-      this.profiles = this.profiles.filter(p => p.id !== id)
-      if (this.activeId === id) this.activeId = this.profiles[0]?.id || ''
-      await this.persist()
+      try {
+        const res = await VoiceService.delete(id)
+        
+        if (res.code === 0) {
+          this.profiles = this.profiles.filter(p => p.id !== id)
+          if (this.activeId === id) {
+            this.activeId = this.profiles[0]?.id || ''
+          }
+        } else {
+          message.error(res.msg || '删除失败')
+          throw new Error(res.msg || '删除失败')
+        }
+      } catch (error) {
+        console.error('删除配音失败:', error)
+        throw error
+      }
     },
+
+    /**
+     * 选择配音
+     */
     select(id) {
       this.activeId = id
+    },
+
+    /**
+     * 刷新数据
+     */
+    async refresh() {
+      this.loaded = false
+      await this.load()
     }
   }
 })
-
-
diff --git a/frontend/app/web-gold/src/utils/video-cover.ts b/frontend/app/web-gold/src/utils/video-cover.ts
index 2d84b5793f..8249e87e23 100644
--- a/frontend/app/web-gold/src/utils/video-cover.ts
+++ b/frontend/app/web-gold/src/utils/video-cover.ts
@@ -60,7 +60,7 @@ export function extractVideoCover(
     const url = URL.createObjectURL(file)
     video.src = url
 
-    let timeoutId: NodeJS.Timeout | null = null
+    let timeoutId: ReturnType<typeof setTimeout> | null = null
 
     const cleanup = () => {
       if (timeoutId) {
diff --git a/frontend/app/web-gold/src/views/dh/Avatar.vue b/frontend/app/web-gold/src/views/dh/Avatar.vue
index abdfed046b..cc21574032 100644
--- a/frontend/app/web-gold/src/views/dh/Avatar.vue
+++ b/frontend/app/web-gold/src/views/dh/Avatar.vue
@@ -4,14 +4,14 @@
 <template>
   <div class="space-y-4">
     <h2 class="text-xl font-bold">生成数字人</h2>
-    <div class="grid grid-cols-1 lg:grid-cols-3 gap-4">
-      <section class="bg-white p-4 rounded shadow lg:col-span-1">
+    <div class="grid grid-cols-1 gap-4 lg:grid-cols-3">
+      <section class="p-4 bg-white rounded shadow lg:col-span-1">
         <div class="space-y-3">
-          <div class="text-gray-600 text-sm">形象、背景、脚本、分辨率、字幕等配置。</div>
-          <button class="px-4 py-2 bg-purple-600 text-white rounded">生成视频</button>
+          <div class="text-sm text-gray-600">形象、背景、脚本、分辨率、字幕等配置。</div>
+          <button class="px-4 py-2 text-white bg-purple-600 rounded">生成视频</button>
         </div>
       </section>
-      <section class="bg-white p-4 rounded shadow lg:col-span-2">
+      <section class="p-4 bg-white rounded shadow lg:col-span-2">
         <div class="text-gray-500">视频预览、任务队列、渲染进度</div>
       </section>
     </div>
diff --git a/frontend/app/web-gold/src/views/dh/Video.vue b/frontend/app/web-gold/src/views/dh/Video.vue
index 023240433f..5e84b73c87 100644
--- a/frontend/app/web-gold/src/views/dh/Video.vue
+++ b/frontend/app/web-gold/src/views/dh/Video.vue
@@ -1,21 +1,22 @@
 <script setup>
-import { ref, computed, onMounted } from 'vue'
+defineOptions({
+  name: 'DigitalVideoPage'
+})
+import { ref, computed, onMounted, watch, onUnmounted } from 'vue'
 import { message } from 'ant-design-vue'
 import { InboxOutlined } from '@ant-design/icons-vue'
 import { useVoiceCopyStore } from '@/stores/voiceCopy'
+import { VoiceService } from '@/api/voice'
 
 const voiceStore = useVoiceCopyStore()
 
-// 标签页模式：'tts' | 'upload'
-const activeTab = ref('upload')
-// 音频模式：'upload' | 'select'
-const audioMode = ref('upload')
-const selectedVoiceId = ref('')
-const uploadedAudio = ref('') // 上传的音频 dataURL 或 URL
 const uploadedVideo = ref('') // 上传的视频 dataURL 或 URL
 const previewVideoUrl = ref('') // 生成的预览视频 URL
 const isGenerating = ref(false)
 const generationProgress = ref(0)
+const isSynthesizing = ref(false)
+const synthesizedAudio = ref(null)
+const previewLoadingVoiceId = ref('')
 
 // 文本转语音相关
 const ttsText = ref('')
@@ -23,19 +24,171 @@ const selectedTtsVoice = ref('')
 const speechRate = ref(1.0)
 const emotion = ref('neutral')
 
-// 推荐声音列表（从已有声音中选择）
-const recommendedVoices = computed(() => voiceStore.profiles.slice(0, 6))
+
+
+const SYSTEM_VOICE_LIBRARY = [
+  {
+    id: 'sys-pro-01',
+    name: '星悦·知性女声',
+    gender: 'female',
+    category: '职业',
+    description: '温柔专业，适合解说、主持场景',
+    previewUrl:
+      'https://cdn.pixabay.com/download/audio/2021/09/19/audio_9b73e92f72.mp3?filename=female-narrator-8863.mp3',
+    source: 'system',
+    voiceId: 'cosyvoice-v2-sys-pro-01',
+    model: 'cosyvoice-v2'
+  },
+  {
+    id: 'sys-boy-01',
+    name: '澄澄·少男音',
+    gender: 'male',
+    category: '少男',
+    description: '年轻清爽，适合青春、校园题材',
+    previewUrl:
+      'https://cdn.pixabay.com/download/audio/2023/04/12/audio_7109b6a4b0.mp3?filename=youthful-voice-143376.mp3',
+    source: 'system',
+    voiceId: 'cosyvoice-v2-sys-boy-01',
+    model: 'cosyvoice-v2'
+  },
+  {
+    id: 'sys-girl-01',
+    name: '沁雪·少女音',
+    gender: 'female',
+    category: '少女',
+    description: '活泼甜美，适合时尚、电商视频',
+    previewUrl:
+      'https://cdn.pixabay.com/download/audio/2021/10/24/audio_b6eed47c13.mp3?filename=teenage-girl-talk-11051.mp3',
+    source: 'system',
+    voiceId: 'cosyvoice-v2-sys-girl-01',
+    model: 'cosyvoice-v2'
+  },
+  {
+    id: 'sys-man-01',
+    name: '寰宇·男青年',
+    gender: 'male',
+    category: '男青年',
+    description: '磁性沉稳，适合宣传片、纪录片',
+    previewUrl:
+      'https://cdn.pixabay.com/download/audio/2022/03/09/audio_e96ada2c9f.mp3?filename=male-voice-10479.mp3',
+    source: 'system',
+    voiceId: 'cosyvoice-v2-sys-man-01',
+    model: 'cosyvoice-v2'
+  }
+]
+
+const voiceSource = ref('system')
+
+const userVoiceCards = computed(() =>
+  (voiceStore.profiles || []).map((profile) => ({
+    id: `user-${profile.id}`,
+    rawId: profile.id,
+    name: profile.name || '未命名',
+    category: profile.gender === 'male' ? '男青年' : '女青',
+    gender: profile.gender || 'female',
+    description: profile.note || '我的配音',
+    previewUrl: profile.fileUrl,
+    fileUrl: profile.fileUrl,
+    transcription: profile.transcription || '',
+    source: 'user',
+    voiceId: profile.voiceId
+  }))
+)
+
+const displayedVoices = computed(() =>
+  voiceSource.value === 'system' ? SYSTEM_VOICE_LIBRARY : userVoiceCards.value
+)
+
+const selectedVoiceMeta = computed(() => {
+  if (!selectedTtsVoice.value) return null
+  return (displayedVoices.value || []).find(
+    (voice) => `${voice.source}-${voice.id}` === selectedTtsVoice.value
+  ) || null
+})
+
+const speechRateMarks = {
+  0.5: '0.5x',
+  1: '1x',
+  1.5: '1.5x',
+  2: '2x'
+}
+
+const speechRateDisplay = computed(() => `${speechRate.value.toFixed(1)}x`)
+
+const setVoiceSource = (source) => {
+  if (voiceSource.value === source) return
+  voiceSource.value = source
+  if (source === 'user' && userVoiceCards.value.length === 0) {
+    selectedTtsVoice.value = ''
+  }
+}
+
+const selectVoiceProfile = (voice) => {
+  selectedTtsVoice.value = `${voice.source}-${voice.id}`
+}
+
+const playVoiceSample = async (voice) => {
+  if (!voice) return
+  if (voice.source === 'user' && voice.voiceId) {
+    return triggerVoicePreview(voice, voice.voiceId)
+  }
+  if (voice.source === 'user' && voice.fileUrl) {
+    return triggerVoicePreview(voice, '')
+  }
+  const url = voice.previewUrl || voice.referenceAudio || voice.fileUrl
+  if (!url) {
+    message.warning('暂无可试听的音频')
+    return
+  }
+  playAudioPreview(url)
+}
+
+async function triggerVoicePreview(voice, voiceId) {
+  previewLoadingVoiceId.value = voice.id
+  try {
+    const res = await VoiceService.preview({
+      voiceId: voiceId || undefined,
+      fileUrl: voice.fileUrl || voice.previewUrl,
+      transcriptionText: voice.transcription || '',
+      inputText: ttsText.value || '',
+      emotion: emotion.value,
+      speechRate: speechRate.value,
+      audioFormat: 'mp3'
+    })
+    if (res.code === 0 && res.data?.audioUrl) {
+      playAudioPreview(res.data.audioUrl)
+    } else if (res.code === 0 && res.data?.audioBase64) {
+      playAudioFromBase64(res.data.audioBase64, res.data.format)
+    } else {
+      message.error(res.msg || '试听失败，请稍后再试')
+    }
+  } catch (error) {
+    console.error('preview voice error:', error)
+    message.error('试听失败，请稍后再试')
+  } finally {
+    previewLoadingVoiceId.value = ''
+  }
+}
+
+const resetSpeechRate = () => {
+  speechRate.value = 1
+}
+
+const ensureDefaultVoice = () => {
+  if (voiceSource.value === 'system' && !selectedTtsVoice.value && SYSTEM_VOICE_LIBRARY.length > 0) {
+    selectedTtsVoice.value = `system-${SYSTEM_VOICE_LIBRARY[0].id}`
+  } else if (
+    voiceSource.value === 'user' &&
+    !selectedTtsVoice.value &&
+    userVoiceCards.value.length > 0
+  ) {
+    selectVoiceProfile(userVoiceCards.value[0])
+  }
+}
 
 // 检查是否可以生成
 const canGenerate = computed(() => {
-  let hasAudio = false
-  if (activeTab.value === 'tts') {
-    hasAudio = !!ttsText.value.trim()
-  } else if (audioMode.value === 'upload') {
-    hasAudio = !!uploadedAudio.value
-  } else {
-    hasAudio = !!selectedVoiceId.value
-  }
+  const hasAudio = !!(synthesizedAudio.value && synthesizedAudio.value.fileId)
   const hasVideo = !!uploadedVideo.value
   return hasAudio && hasVideo && !isGenerating.value
 })
@@ -50,50 +203,6 @@ function toDataURL(file) {
   })
 }
 
-// 音频上传处理
-async function handleAudioUpload(file) {
-  // 检查文件大小（50MB）
-  if (file.size > 50 * 1024 * 1024) {
-    message.error('文件大小不能超过 50MB')
-    return false
-  }
-  
-  // 检查文件类型
-  if (!file.type.startsWith('audio/')) {
-    message.error('请上传音频文件')
-    return false
-  }
-  
-  try {
-    const dataUrl = await toDataURL(file)
-    uploadedAudio.value = dataUrl
-    message.success('音频上传成功')
-  } catch (error) {
-    message.error('音频上传失败，请重试')
-    console.error(error)
-  }
-  return false // 阻止自动上传
-}
-
-// 音频拖拽处理（a-upload-dragger 会自动处理，这里只处理粘贴）
-async function handleAudioDrop(e) {
-  // a-upload-dragger 会自动处理拖拽，这里不需要额外处理
-  console.log('Audio drop event:', e)
-}
-
-// 音频粘贴处理
-function handleAudioPaste(e) {
-  const items = Array.from(e.clipboardData.items)
-  const audioItem = items.find(item => item.type.startsWith('audio/'))
-  
-  if (audioItem) {
-    const file = audioItem.getAsFile()
-    if (file) {
-      handleAudioUpload(file)
-    }
-  }
-}
-
 // 视频上传处理
 async function handleVideoUpload(file) {
   // 检查文件类型
@@ -113,18 +222,60 @@ async function handleVideoUpload(file) {
   return false // 阻止自动上传
 }
 
+async function handleSynthesizeVoice() {
+  if (!ttsText.value.trim()) {
+    message.warning('请输入你想让角色说的话')
+    return
+  }
+  const voice = selectedVoiceMeta.value
+  if (!voice) {
+    message.warning('请选择音色')
+    return
+  }
+  const voiceId = voice.voiceId || voice.rawId || voice.id
+  if (!voiceId) {
+    message.warning('该音色暂未配置 voiceId，无法合成')
+    return
+  }
+  isSynthesizing.value = true
+  try {
+    const res = await VoiceService.synthesize({
+      inputText: ttsText.value,
+      transcriptionText: selectedVoiceMeta.value?.transcription || '',
+      voiceId,
+      model: voice.model || undefined,
+      speechRate: speechRate.value,
+      emotion: emotion.value,
+      audioFormat: 'mp3'
+    })
+    if (res.code === 0) {
+      synthesizedAudio.value = res.data
+      message.success('语音合成成功，可用于生成视频')
+    } else {
+      message.error(res.msg || '语音合成失败，请稍后重试')
+    }
+  } catch (error) {
+    console.error('synthesize voice error:', error)
+    message.error('语音合成失败，请稍后再试')
+  } finally {
+    isSynthesizing.value = false
+  }
+}
+
+function playSynthesizedAudio() {
+  if (!synthesizedAudio.value?.audioUrl) {
+    message.warning('请先生成语音')
+    return
+  }
+  playAudioPreview(synthesizedAudio.value.audioUrl)
+}
+
 // 视频拖拽处理（a-upload-dragger 会自动处理）
 async function handleVideoDrop(e) {
   // a-upload-dragger 会自动处理拖拽，这里不需要额外处理
   console.log('Video drop event:', e)
 }
 
-// 切换标签页
-function switchTab(tab) {
-  activeTab.value = tab
-}
-
-
 // 生成视频
 async function generateVideo() {
   if (!canGenerate.value) {
@@ -161,12 +312,6 @@ async function generateVideo() {
   }
 }
 
-// 清除音频
-function clearAudio() {
-  uploadedAudio.value = ''
-  message.info('已清除音频')
-}
-
 // 清除视频
 function clearVideo() {
   uploadedVideo.value = ''
@@ -188,21 +333,91 @@ function downloadPreview() {
 }
 
 // 播放音频预览
-function playAudioPreview(url) {
+function playAudioPreview(url, options = {}) {
+  if (!url) {
+    message.warning('暂无可试听的音频')
+    return
+  }
+  try {
+    previewAudio?.pause?.()
+  } catch {
+    // ignore
+  }
   const audio = new Audio(url)
-  audio.play().catch(err => {
-    console.error('播放失败:', err)
-    message.error('播放失败')
-  })
+  audio.play()
+    .then(() => {
+      previewAudio = audio
+      audio.onended = () => {
+        if (options.revokeOnEnd && url.startsWith('blob:')) {
+          URL.revokeObjectURL(url)
+          if (previewObjectUrl === url) {
+            previewObjectUrl = ''
+          }
+        }
+      }
+    })
+    .catch(err => {
+      console.error('播放失败:', err)
+      message.error('播放失败，请稍后重试')
+    })
 }
 
+function playAudioFromBase64(audioBase64, format = 'mp3') {
+  try {
+    if (previewObjectUrl) {
+      URL.revokeObjectURL(previewObjectUrl)
+      previewObjectUrl = ''
+    }
+    const byteCharacters = window.atob(audioBase64)
+    const byteNumbers = new Array(byteCharacters.length)
+    for (let i = 0; i < byteCharacters.length; i += 1) {
+      byteNumbers[i] = byteCharacters.charCodeAt(i)
+    }
+    const mime = format === 'mp3' ? 'audio/mpeg' : `audio/${format}`
+    const blob = new Blob([new Uint8Array(byteNumbers)], { type: mime })
+    previewObjectUrl = URL.createObjectURL(blob)
+    playAudioPreview(previewObjectUrl, { revokeOnEnd: true })
+  } catch (error) {
+    console.error('playAudioFromBase64 failed:', error)
+    message.error('音频播放失败，请稍后重试')
+  }
+}
+
+let previewAudio = null
+let previewObjectUrl = ''
+
 // 初始化：加载已有声音列表
 onMounted(async () => {
   await voiceStore.load()
-  if (voiceStore.profiles.length > 0 && !selectedVoiceId.value) {
-    selectedVoiceId.value = voiceStore.activeId || voiceStore.profiles[0]?.id || ''
+  ensureDefaultVoice()
+})
+
+onUnmounted(() => {
+  previewAudio?.pause?.()
+  previewAudio = null
+  if (previewObjectUrl) {
+    URL.revokeObjectURL(previewObjectUrl)
+    previewObjectUrl = ''
   }
 })
+
+watch(voiceSource, () => {
+  selectedTtsVoice.value = ''
+  ensureDefaultVoice()
+})
+
+watch(
+  () => voiceStore.profiles,
+  () => {
+    if (voiceSource.value === 'user') {
+      ensureDefaultVoice()
+  }
+  }
+)
+
+watch([ttsText, selectedTtsVoice], () => {
+  synthesizedAudio.value = null
+})
 </script>
 
 <template>
@@ -210,26 +425,6 @@ onMounted(async () => {
     <div class="digital-video-grid">
       <!-- 左侧：配置区域 -->
       <section class="digital-video-left">
-        <!-- 标签页 -->
-        <div class="tab-container">
-          <button 
-            class="tab-btn" 
-            :class="{ active: activeTab === 'tts' }"
-            @click="switchTab('tts')"
-          >
-            文本朗读
-          </button>
-          <button 
-            class="tab-btn" 
-            :class="{ active: activeTab === 'upload' }"
-            @click="switchTab('upload')"
-          >
-            上传本地配音
-          </button>
-        </div>
-
-        <!-- 文本朗读标签页内容 -->
-        <div v-if="activeTab === 'tts'" class="tab-content">
           <div class="tts-section">
             <div class="tts-textarea-wrapper">
               <a-textarea 
@@ -244,44 +439,81 @@ onMounted(async () => {
               </div>
             </div>
 
-            <!-- 音色选择 -->
             <div class="voice-selection-section">
               <div class="section-label">音色</div>
-              <div class="voice-categories">
-                <span class="category-tag active">全部</span>
-                <span class="category-tag">职业</span>
-                <span class="category-tag">少男</span>
-                <span class="category-tag">少女</span>
-                <span class="category-tag">男青年</span>
-                <span class="category-tag">女青</span>
+
+            <div class="voice-source-toggle">
+              <button
+                v-for="source in ['system', 'user']"
+                :key="source"
+                class="source-btn"
+                :class="{ active: voiceSource === source }"
+                @click="setVoiceSource(source)"
+              >
+                {{ source === 'system' ? '系统音色' : '我的音色' }}
+              </button>
               </div>
-              <div class="voice-profiles">
+
+            <div
+              v-if="voiceSource === 'user' && userVoiceCards.length === 0"
+              class="empty-voices"
+            >
+              还没有配音，可先在“配音管理”中上传
+            </div>
+
+            <div class="voice-list">
                 <div 
-                  v-for="voice in recommendedVoices" 
+                v-for="voice in displayedVoices"
                   :key="voice.id"
-                  class="voice-profile-card"
-                  @click="selectedTtsVoice = voice.id"
+                class="voice-list-item"
+                :class="{ selected: selectedTtsVoice === `${voice.source}-${voice.id}` }"
+                @click="selectVoiceProfile(voice)"
                 >
-                  <div class="voice-avatar">👤</div>
-                  <div class="voice-name">{{ voice.name || '未命名' }}</div>
-                  <button class="play-btn" @click.stop="playAudioPreview(voice.referenceAudio)">
-                    ▶
-                  </button>
+                <div class="voice-item-info">
+                  <div class="voice-avatar">{{ voice.gender === 'male' ? '👨' : '👩' }}</div>
+                  <div class="voice-item-text">
+                    <div class="voice-name">{{ voice.name }}</div>
+                    <div class="voice-desc">{{ voice.description }}</div>
+                  </div>
+                </div>
+                <div class="voice-item-meta">
+                  <span class="voice-tag">{{ voice.category }}</span>
+                  <span class="voice-tag" :class="{ system: voice.source === 'system' }">
+                    {{ voice.source === 'system' ? '系统' : '我的' }}
+                  </span>
+                  <a-button
+                    size="small"
+                    type="link"
+                    :loading="previewLoadingVoiceId === voice.id"
+                    @click.stop="playVoiceSample(voice)"
+                  >
+                    试听
+                  </a-button>
+                </div>
                 </div>
               </div>
             </div>
 
-            <!-- 语速和情感 -->
-            <div class="tts-controls">
-              <div class="control-group">
-                <div class="control-label">语速</div>
+          <div class="tts-controls">
+            <div class="control-group">
+              <div class="control-label">语速</div>
+              <div class="slider-card">
+                <div class="slider-info">
+                  <div class="slider-value">{{ speechRateDisplay }}</div>
+                  <button class="reset-btn" @click="resetSpeechRate">重置</button>
+                </div>
                 <div class="slider-control">
-                  <button class="slider-btn">◀</button>
-                  <a-slider v-model:value="speechRate" :min="0.5" :max="2" :step="0.1" />
-                  <button class="slider-btn">▶</button>
-                  <button class="reset-btn">1x</button>
+                  <a-slider
+                    v-model:value="speechRate"
+                    :min="0.5"
+                    :max="2"
+                    :step="0.1"
+                    :marks="speechRateMarks"
+                    :tooltip-open="false"
+                  />
                 </div>
               </div>
+            </div>
               <div class="control-group">
                 <div class="control-label">情感</div>
                 <div class="emotion-buttons">
@@ -294,56 +526,39 @@ onMounted(async () => {
                   >
                     {{ em === 'neutral' ? '中性' : em === 'happy' ? '高兴' : em === 'angry' ? '愤怒' : em === 'sad' ? '悲伤' : em === 'scared' ? '害怕' : em === 'disgusted' ? '厌恶' : '惊讶' }}
                   </button>
-                </div>
               </div>
             </div>
           </div>
         </div>
 
-        <!-- 上传本地配音标签页内容 -->
-        <div v-if="activeTab === 'upload'" class="tab-content">
-          <!-- 上传区域 -->
-          <div class="upload-section">
-        
-            <a-upload-dragger
-              name="file"
-              accept="audio/*"
-              action=""
-              :before-upload="handleAudioUpload"
-              :show-upload-list="false"
-              @drop="handleAudioDrop"
-              @paste="handleAudioPaste"
+        <div class="video-section">
+          <div class="tts-actions">
+            <a-button
+              type="primary"
+              size="large"
+              block
+              :loading="isSynthesizing"
+              @click="handleSynthesizeVoice"
             >
-              <p class="ant-upload-drag-icon">
-                <InboxOutlined />
-              </p>
-              <p class="ant-upload-text">点击或拖拽音频文件到此处上传</p>
-              <p class="ant-upload-hint">
-                支持点击、拖拽或粘贴上传
-                <br />
-                支持格式：MP3、WAV、M4A、FLAC、AAC、OGG
-                <br />
-                文件大小 ≤ 50MB，时长 ≤ 60 秒
-              </p>
-            </a-upload-dragger>
-            
-            <div v-if="uploadedAudio" class="uploaded-audio-preview">
-              <div class="preview-header">
-                <span class="preview-label">已上传音频</span>
-                <a-button type="link" danger @click="clearAudio" size="small">
-                  清除
+              {{ isSynthesizing ? '语音合成中...' : '生成语音' }}
+            </a-button>
+            <div v-if="synthesizedAudio" class="synth-audio-card">
+              <div class="synth-audio-title">已生成语音</div>
+              <div class="synth-audio-meta">
+                <span>文件编号：{{ synthesizedAudio.fileId }}</span>
+                <span>格式：{{ (synthesizedAudio.format || 'mp3').toUpperCase() }}</span>
+              </div>
+              <div class="synth-audio-actions">
+                <a-button type="default" size="small" @click="playSynthesizedAudio">
+                  试听生成语音
                 </a-button>
               </div>
-              <audio :src="uploadedAudio" controls preload="metadata" />
             </div>
-      
+            <div v-else class="synth-audio-hint">
+              先生成语音，再上传或匹配视频，即可开始混剪
+            </div>
           </div>
 
- 
-        </div>
-
-        <!-- 视频配置 -->
-        <div class="video-section">
           <div class="section-label">视频</div>
  
           <a-upload-dragger
@@ -376,10 +591,8 @@ onMounted(async () => {
             </div>
             <video :src="uploadedVideo" controls preload="metadata" />
           </div>
-
         </div>
 
-        <!-- 生成按钮 -->
         <div class="generate-section">
           <a-button 
             type="primary" 
@@ -490,53 +703,6 @@ onMounted(async () => {
   background: rgba(59, 130, 246, 0.5);
 }
 
-/* 标签页 */
-.tab-container {
-  display: flex;
-  gap: 4px;
-  margin-bottom: 16px;
-  background: rgba(0, 0, 0, 0.2);
-  padding: 4px;
-  border-radius: 8px;
-  border: 1px solid rgba(59, 130, 246, 0.1);
-}
-
-.tab-btn {
-  flex: 1;
-  padding: 10px 20px;
-  border: none;
-  background: transparent;
-  color: var(--color-text-secondary);
-  font-size: 14px;
-  font-weight: 500;
-  cursor: pointer;
-  border-radius: 8px;
-  transition: all 0.3s cubic-bezier(0.4, 0, 0.2, 1);
-  position: relative;
-}
-
-.tab-btn:hover {
-  color: var(--color-text);
-  background: rgba(255, 255, 255, 0.03);
-}
-
-.tab-btn.active {
-  color: var(--color-text);
-  background: linear-gradient(135deg, rgba(59, 130, 246, 0.2) 0%, rgba(139, 92, 246, 0.15) 100%);
-  box-shadow: 
-    0 2px 8px rgba(59, 130, 246, 0.2),
-    inset 0 1px 0 rgba(255, 255, 255, 0.1);
-  border: 1px solid rgba(59, 130, 246, 0.3);
-}
-
-.tab-content {
-  margin-bottom: 16px;
-  min-height: 400px;
-  flex: 1;
-  display: flex;
-  flex-direction: column;
-}
-
 /* 文本朗读区域 */
 .tts-section {
   display: flex;
@@ -608,75 +774,71 @@ onMounted(async () => {
   padding-bottom: 4px;
 }
 
-.category-tag {
-  padding: 8px 16px;
+.voice-source-toggle {
+  display: inline-flex;
   border: 1px solid rgba(59, 130, 246, 0.2);
-  border-radius: 20px;
-  font-size: 12px;
-  color: var(--color-text-secondary);
-  background: rgba(0, 0, 0, 0.3);
-  cursor: pointer;
-  white-space: nowrap;
-  transition: all 0.3s cubic-bezier(0.4, 0, 0.2, 1);
-  font-weight: 500;
-}
-
-.category-tag:hover {
-  border-color: rgba(59, 130, 246, 0.5);
-  color: var(--color-text);
-  background: rgba(59, 130, 246, 0.1);
-  transform: translateY(-1px);
-}
-
-.category-tag.active {
-  background: linear-gradient(135deg, var(--color-primary) 0%, rgba(139, 92, 246, 0.8) 100%);
-  border-color: var(--color-primary);
-  color: white;
-  box-shadow: 0 2px 8px rgba(59, 130, 246, 0.3);
-  transform: translateY(-1px);
-}
-
-.voice-profiles {
-  display: grid;
-  grid-template-columns: repeat(3, 1fr);
-  gap: 8px;
-}
-
-.voice-profile-card {
-  display: flex;
-  flex-direction: column;
-  align-items: center;
-  padding: 12px;
-  background: linear-gradient(135deg, rgba(0, 0, 0, 0.4) 0%, rgba(26, 26, 26, 0.3) 100%);
-  border: 1px solid rgba(59, 130, 246, 0.15);
-  border-radius: var(--radius-card);
-  cursor: pointer;
-  transition: all 0.3s cubic-bezier(0.4, 0, 0.2, 1);
-  position: relative;
+  border-radius: 24px;
   overflow: hidden;
 }
 
-.voice-profile-card::before {
-  content: '';
-  position: absolute;
-  top: 0;
-  left: 0;
-  right: 0;
-  bottom: 0;
-  background: linear-gradient(135deg, rgba(59, 130, 246, 0.1) 0%, rgba(139, 92, 246, 0.05) 100%);
-  opacity: 0;
-  transition: opacity 0.3s ease;
+.source-btn {
+  padding: 6px 16px;
+  font-size: 12px;
+  border: none;
+  background: transparent;
+  color: var(--color-text-secondary);
+  cursor: pointer;
+  transition: all 0.2s ease;
 }
 
-.voice-profile-card:hover {
+.source-btn.active {
+  background: rgba(59, 130, 246, 0.2);
+  color: var(--color-text);
+}
+
+.empty-voices {
+  padding: 8px 12px;
+  font-size: 12px;
+  color: var(--color-text-secondary);
+  background: rgba(0, 0, 0, 0.3);
+  border: 1px dashed rgba(59, 130, 246, 0.3);
+  border-radius: var(--radius-card);
+}
+
+.voice-list {
+  display: flex;
+  flex-direction: column;
+  gap: 10px;
+  margin-top: 8px;
+}
+
+.voice-list-item {
+  display: flex;
+  justify-content: space-between;
+  align-items: center;
+  padding: 12px 14px;
+  border: 1px solid rgba(59, 130, 246, 0.15);
+  border-radius: 10px;
+  background: rgba(0, 0, 0, 0.35);
+  cursor: pointer;
+  transition: all 0.2s ease;
+}
+
+.voice-list-item:hover {
   border-color: rgba(59, 130, 246, 0.5);
-  background: linear-gradient(135deg, rgba(59, 130, 246, 0.15) 0%, rgba(139, 92, 246, 0.1) 100%);
-  transform: translateY(-2px);
-  box-shadow: 0 4px 12px rgba(59, 130, 246, 0.2);
+  background: rgba(59, 130, 246, 0.08);
 }
 
-.voice-profile-card:hover::before {
-  opacity: 1;
+.voice-list-item.selected {
+  border-color: var(--color-primary);
+  box-shadow: 0 0 0 1px rgba(59, 130, 246, 0.4);
+  background: linear-gradient(135deg, rgba(59, 130, 246, 0.2) 0%, rgba(139, 92, 246, 0.15) 100%);
+}
+
+.voice-item-info {
+  display: flex;
+  align-items: center;
+  gap: 10px;
 }
 
 .voice-avatar {
@@ -688,33 +850,43 @@ onMounted(async () => {
   align-items: center;
   justify-content: center;
   font-size: 20px;
-  margin-bottom: 8px;
+  flex-shrink: 0;
+}
+
+.voice-item-text {
+  display: flex;
+  flex-direction: column;
+  gap: 2px;
 }
 
 .voice-name {
-  font-size: 12px;
+  font-size: 13px;
   color: var(--color-text);
-  text-align: center;
-  margin-bottom: 4px;
+  font-weight: 600;
 }
 
-.play-btn {
-  width: 24px;
-  height: 24px;
-  border-radius: 50%;
-  border: none;
-  background: var(--color-primary);
-  color: white;
+.voice-desc {
+  font-size: 12px;
+  color: var(--color-text-secondary);
+}
+
+.voice-item-meta {
   display: flex;
   align-items: center;
-  justify-content: center;
-  font-size: 10px;
-  cursor: pointer;
-  transition: all 0.2s ease;
+  gap: 8px;
 }
 
-.play-btn:hover {
-  transform: scale(1.1);
+.voice-tag {
+  border: 1px solid rgba(255, 255, 255, 0.25);
+  border-radius: 20px;
+  padding: 2px 10px;
+  font-size: 11px;
+  color: var(--color-text-secondary);
+}
+
+.voice-tag.system {
+  color: var(--color-primary);
+  border-color: rgba(59, 130, 246, 0.5);
 }
 
 /* TTS 控制 */
@@ -737,29 +909,31 @@ onMounted(async () => {
 }
 
 .slider-control {
+  width: 100%;
+}
+
+.slider-card {
+  border: 1px solid rgba(59, 130, 246, 0.2);
+  border-radius: 12px;
+  padding: 10px 12px;
+  background: rgba(0, 0, 0, 0.25);
   display: flex;
-  align-items: center;
+  flex-direction: column;
   gap: 8px;
 }
 
-.slider-btn {
-  width: 28px;
-  height: 28px;
-  border: 1px solid var(--color-border);
-  background: var(--color-surface);
-  color: var(--color-text);
-  border-radius: 4px;
-  cursor: pointer;
+.slider-info {
   display: flex;
   align-items: center;
-  justify-content: center;
+  justify-content: space-between;
   font-size: 12px;
-  transition: all 0.2s ease;
+  color: var(--color-text-secondary);
 }
 
-.slider-btn:hover {
-  border-color: var(--color-primary);
-  background: rgba(59, 130, 246, 0.1);
+.slider-value {
+  font-size: 14px;
+  font-weight: 600;
+  color: var(--color-text);
 }
 
 .reset-btn {
@@ -812,172 +986,48 @@ onMounted(async () => {
   transform: translateY(-1px);
 }
 
-/* 上传区域 */
-.upload-section {
-  margin-bottom: 16px;
-  flex: 1;
+.tts-actions {
+  margin: 16px 0;
   display: flex;
   flex-direction: column;
   gap: 12px;
 }
 
-.upload-hint-box {
-  display: flex;
-  align-items: center;
-  gap: 8px;
-  padding: 10px 12px;
+.synth-audio-card {
   background: rgba(59, 130, 246, 0.08);
-  border: 1px solid rgba(59, 130, 246, 0.2);
+  border: 1px solid rgba(59, 130, 246, 0.3);
   border-radius: var(--radius-card);
-  margin-bottom: 4px;
-}
-
-.hint-icon {
-  font-size: 16px;
-  flex-shrink: 0;
-}
-
-.hint-text {
-  font-size: 12px;
-  color: var(--color-text-secondary);
-  line-height: 1.4;
-}
-
-.upload-wrapper {
-  width: 100%;
-}
-
-.upload-zone {
-  width: 100%;
-  min-height: 200px;
-  border: 1.5px dashed rgba(59, 130, 246, 0.4);
-  border-radius: 8px;
-  padding: 40px 20px;
-  text-align: center;
-  cursor: pointer;
-  transition: all 0.3s cubic-bezier(0.4, 0, 0.2, 1);
-  background: rgba(0, 0, 0, 0.2);
-  position: relative;
-  overflow: hidden;
-}
-
-.upload-zone-content {
-  display: flex;
-  flex-direction: column;
-  align-items: center;
-  justify-content: center;
-  gap: 16px;
-  position: relative;
-  z-index: 1;
-}
-
-.upload-icon-wrapper {
-  width: 64px;
-  height: 64px;
-  display: flex;
-  align-items: center;
-  justify-content: center;
-  border-radius: 50%;
-  background: rgba(59, 130, 246, 0.1);
-  margin-bottom: 4px;
-}
-
-.upload-icon-svg {
-  width: 32px;
-  height: 32px;
-  color: var(--color-primary);
-  stroke-width: 1.5;
-}
-
-.upload-zone::before {
-  content: '';
-  position: absolute;
-  top: 0;
-  left: 0;
-  right: 0;
-  bottom: 0;
-  background: linear-gradient(135deg, rgba(59, 130, 246, 0.1) 0%, rgba(139, 92, 246, 0.05) 100%);
-  opacity: 0;
-  transition: opacity 0.3s ease;
-}
-
-.upload-zone:hover {
-  border-color: var(--color-primary);
-  background: rgba(59, 130, 246, 0.08);
-  border-style: solid;
-}
-
-.upload-zone:hover .upload-icon-wrapper {
-  background: rgba(59, 130, 246, 0.15);
-  transform: scale(1.05);
-}
-
-.upload-zone.dragging {
-  border-color: var(--color-primary);
-  background: rgba(59, 130, 246, 0.12);
-  border-style: solid;
-  border-width: 2px;
-}
-
-.upload-zone.dragging .upload-icon-wrapper {
-  background: rgba(59, 130, 246, 0.2);
-  transform: scale(1.1);
-}
-
-.upload-main-text {
-  font-size: 16px;
-  font-weight: 500;
+  padding: 12px;
   color: var(--color-text);
-  margin-top: 4px;
 }
 
-.upload-sub-text {
-  font-size: 14px;
-  color: var(--color-text-secondary);
-  font-weight: 400;
+.synth-audio-title {
+  font-size: 13px;
+  font-weight: 600;
+  margin-bottom: 4px;
 }
 
-.format-tags {
-  display: flex;
-  flex-wrap: wrap;
-  gap: 6px;
-  justify-content: center;
-  margin-top: 4px;
-}
-
-.format-tag {
-  display: inline-block;
-  padding: 4px 10px;
+.synth-audio-meta {
   font-size: 12px;
-  font-weight: 500;
-  color: var(--color-primary);
-  background: rgba(59, 130, 246, 0.1);
-  border: 1px solid rgba(59, 130, 246, 0.2);
-  border-radius: 4px;
-  line-height: 1.2;
-}
-
-.upload-limit-text {
-  font-size: 12px;
-  color: var(--color-text-secondary);
+  opacity: 0.85;
   display: flex;
-  align-items: center;
-  gap: 8px;
-  flex-wrap: wrap;
-  justify-content: center;
-  margin-top: 4px;
+  justify-content: space-between;
 }
 
-.limit-item {
-  color: var(--color-text-secondary);
+.synth-audio-actions {
+  margin-top: 8px;
+  display: flex;
+  justify-content: flex-end;
 }
 
-.limit-separator {
-  opacity: 0.4;
-  font-size: 10px;
+.synth-audio-hint {
+  font-size: 12px;
+  color: var(--color-text-3);
+  text-align: center;
 }
 
-.uploaded-audio-preview,
+/* 上传区域 */
+
 .uploaded-video-preview {
   padding: 12px;
   background: var(--color-bg);
@@ -998,7 +1048,6 @@ onMounted(async () => {
   color: var(--color-text);
 }
 
-.uploaded-audio-preview audio,
 .uploaded-video-preview video {
   width: 100%;
   margin-bottom: 8px;
diff --git a/frontend/app/web-gold/src/views/dh/VoiceCopy.vue b/frontend/app/web-gold/src/views/dh/VoiceCopy.vue
index 456c243c67..d8398e5ce7 100644
--- a/frontend/app/web-gold/src/views/dh/VoiceCopy.vue
+++ b/frontend/app/web-gold/src/views/dh/VoiceCopy.vue
@@ -1,1041 +1,649 @@
+<template>
+  <div class="voice-copy-page">
+    <!-- 页面头部 -->
+    <div class="page-header">
+      <h1 class="page-title">配音管理</h1>
+      <a-button type="primary" @click="handleCreate">
+        <template #icon>
+          <PlusOutlined />
+        </template>
+        <span>新建配音</span>
+      </a-button>
+    </div>
+
+    <!-- 搜索栏 -->
+    <div class="search-bar">
+      <a-space>
+        <a-input
+          v-model:value="searchParams.name"
+          placeholder="搜索配音名称"
+          style="width: 250px"
+          allow-clear
+          @press-enter="handleSearch"
+        >
+          <template #prefix>
+            <SearchOutlined />
+          </template>
+        </a-input>
+        <a-button type="primary" @click="handleSearch">查询</a-button>
+        <a-button @click="handleReset">重置</a-button>
+      </a-space>
+    </div>
+
+    <!-- 列表表格 -->
+    <div class="table-container">
+      <a-table
+        :columns="columns"
+        :data-source="voiceList"
+        :loading="loading"
+        :pagination="pagination"
+        row-key="id"
+        @change="handleTableChange"
+      >
+        <template #bodyCell="{ column, record }">
+          <template v-if="column.key === 'name'">
+            <div class="voice-name">{{ record.name || '未命名' }}</div>
+          </template>
+          <template v-else-if="column.key === 'transcription'">
+            <div class="transcription-text">{{ formatTranscription(record.transcription) }}</div>
+          </template>
+          <template v-else-if="column.key === 'createTime'">
+            <span>{{ formatDateTime(record.createTime) }}</span>
+          </template>
+          <template v-else-if="column.key === 'fileUrl'">
+            <a-button type="link" size="small" @click="handlePlayAudio(record)">
+              <template #icon>
+                <PlayCircleOutlined />
+              </template>
+              播放
+            </a-button>
+          </template>
+          <template v-else-if="column.key === 'actions'">
+            <a-space>
+              <a-button type="link" size="small" @click="handleEdit(record)">编辑</a-button>
+              <a-button
+                type="link"
+                size="small"
+                :loading="transcribingId === record.id"
+                :disabled="!!record.transcription"
+                @click="handleTranscribe(record)"
+              >
+                {{ record.transcription ? '已识别' : '识别' }}
+              </a-button>
+              <a-button type="link" size="small" danger @click="handleDelete(record)">删除</a-button>
+            </a-space>
+          </template>
+        </template>
+      </a-table>
+    </div>
+
+    <!-- 新建/编辑表单 Modal -->
+    <a-modal
+      v-model:open="modalVisible"
+      :title="isCreateMode ? '新建配音' : '编辑配音'"
+      :width="600"
+      :confirm-loading="submitting"
+      @ok="handleSubmit"
+      @cancel="handleCancel"
+    >
+      <a-form
+        ref="formRef"
+        :model="formData"
+        :rules="formRules"
+        layout="vertical"
+      >
+        <a-form-item label="配音名称" name="name">
+          <a-input v-model:value="formData.name" placeholder="请输入配音名称" />
+        </a-form-item>
+
+        <a-form-item
+          v-if="isCreateMode"
+          label="音频文件"
+          name="fileId"
+          :rules="[{ required: true, message: '请上传音频文件' }]"
+        >
+          <a-upload
+            v-model:file-list="fileList"
+            :custom-request="handleCustomUpload"
+            :before-upload="handleBeforeUpload"
+            :show-upload-list="true"
+            :max-count="1"
+            accept="audio/*,.mp3,.wav,.aac,.m4a,.flac,.ogg"
+            @remove="handleRemoveFile"
+            @change="handleFileListChange"
+          >
+            <a-button type="primary" :loading="uploading">
+              <template #icon>
+                <UploadOutlined v-if="!uploading" />
+              </template>
+              {{ uploading ? '上传中...' : (fileList.length > 0 ? '重新上传' : '上传音频文件') }}
+            </a-button>
+          </a-upload>
+          <div class="upload-hint">支持格式：MP3、WAV、AAC、M4A、FLAC、OGG，单个文件不超过 100MB</div>
+        </a-form-item>
+
+        <a-form-item label="备注" name="note">
+          <a-textarea
+            v-model:value="formData.note"
+            :rows="3"
+            placeholder="请输入备注信息"
+          />
+        </a-form-item>
+
+        <a-form-item
+          v-if="!isCreateMode"
+          label="识别内容"
+          name="transcription"
+        >
+          <a-textarea
+            v-model:value="formData.transcription"
+            :rows="4"
+            placeholder="识别内容，支持手动修改"
+          />
+        </a-form-item>
+      </a-form>
+    </a-modal>
+
+    <!-- 音频播放器 -->
+    <audio ref="audioPlayer" style="display: none" controls />
+  </div>
+</template>
+
 <script setup>
-import { ref, reactive, computed, onMounted } from 'vue'
+import { ref, reactive, computed, onMounted, onUnmounted, nextTick } from 'vue'
 import { message, Modal } from 'ant-design-vue'
-import { InboxOutlined } from '@ant-design/icons-vue'
-import { useVoiceCopyStore } from '@/stores/voiceCopy'
-import useVoiceText from '@gold/hooks/web/useVoiceText'
+import {
+  PlusOutlined,
+  SearchOutlined,
+  UploadOutlined,
+  PlayCircleOutlined
+} from '@ant-design/icons-vue'
+import { VoiceService } from '@/api/voice'
+import { MaterialService } from '@/api/material'
+import dayjs from 'dayjs'
 
-const store = useVoiceCopyStore()
-const profiles = computed(() => store.profiles)
-const activeId = computed(() => store.activeId)
-const { getVoiceText } = useVoiceText()
+// ========== 常量定义 ==========
+const POLLING_CONFIG = {
+  interval: 10000, // 轮询间隔：10秒
+  maxCount: 30, // 最大轮询次数：30次（5分钟）
+  transcriptionMaxLength: 50 // 识别内容最大显示长度
+}
 
-const form = reactive({
-  id: '',
+const DEFAULT_FORM_DATA = {
+  id: null,
   name: '',
-  language: 'zh-CN', // 简体中文
+  fileId: null,
+  autoTranscribe: true,
+  language: 'zh-CN',
   gender: 'female',
-  referenceAudio: '', // dataURL 或外链
-  sampleText: '今天天气很好，我们一起去公园散步吧。',
-  enhancement: 50,    // 增强强度 0-100（去噪/清晰度）
-  noiseReduction: true,
   note: '',
-  originalText: '',   // 原语音文本
+  transcription: ''
+}
+
+// ========== 响应式数据 ==========
+const loading = ref(false)
+const submitting = ref(false)
+const uploading = ref(false)
+const voiceList = ref([])
+const transcribingId = ref(null)
+const modalVisible = ref(false)
+const formMode = ref('create')
+const formRef = ref(null)
+const audioPlayer = ref(null)
+const fileList = ref([])
+let pollingTimer = null
+
+const searchParams = reactive({
+  name: '',
+  pageNo: 1,
+  pageSize: 10
 })
 
-const isSavingAs = ref(false)
-const saveAsName = ref('')
+const pagination = reactive({
+  current: 1,
+  pageSize: 10,
+  total: 0,
+  showSizeChanger: true,
+  showTotal: (total) => `共 ${total} 条`
+})
 
-async function loadProfiles() { await store.load(); if (store.activeProfile) Object.assign(form, { ...store.activeProfile }) }
+const formData = reactive({ ...DEFAULT_FORM_DATA })
 
-function toDataURL(file) {
-  return new Promise((resolve, reject) => {
-    const reader = new FileReader()
-    reader.onload = () => resolve(reader.result)
-    reader.onerror = reject
-    reader.readAsDataURL(file)
+// ========== 计算属性 ==========
+const isCreateMode = computed(() => formMode.value === 'create')
+
+// ========== 表格配置 ==========
+const columns = [
+  { title: '配音名称', key: 'name', dataIndex: 'name', width: 200 },
+  { title: '识别内容', key: 'transcription', dataIndex: 'transcription', width: 300 },
+  { title: '创建时间', key: 'createTime', dataIndex: 'createTime', width: 180 },
+  { title: '操作', key: 'actions', width: 200, fixed: 'right' }
+]
+
+// ========== 表单验证规则 ==========
+const formRules = {
+  name: [{ required: true, message: '请输入配音名称' }],
+  fileId: [{ required: true, message: '请上传音频文件' }]
+}
+
+// ========== 工具函数 ==========
+const formatTranscription = (transcription) => {
+  if (!transcription) return '未识别'
+  if (transcription.length <= POLLING_CONFIG.transcriptionMaxLength) return transcription
+  return transcription.substring(0, POLLING_CONFIG.transcriptionMaxLength) + '...'
+}
+
+const formatDateTime = (value) => {
+  if (!value) return '-'
+  return dayjs(value).format('YYYY-MM-DD HH:mm:ss')
+}
+
+const fillFormData = (data) => {
+  Object.assign(formData, {
+    id: data.id || null,
+    name: data.name || '',
+    fileId: data.fileId || null,
+    language: data.language || 'zh-CN',
+    gender: data.gender || 'female',
+    note: data.note || '',
+    transcription: data.transcription || ''
   })
 }
 
-async function onUploadBefore(file) {
-  // 检查文件大小（50MB）
-  if (file.size > 50 * 1024 * 1024) {
-    message.error('文件大小不能超过 50MB')
-    return false
-  }
-  
-  // 检查文件类型
-  if (!file.type.startsWith('audio/')) {
-    message.error('请上传音频文件')
-    return false
-  }
-  
+// ========== 数据加载 ==========
+const loadVoiceList = async () => {
+  loading.value = true
   try {
-    const dataUrl = await toDataURL(file)
-    form.referenceAudio = dataUrl
-    message.success('音频已就绪（本地预处理）')
-  } catch {
-    message.error('读取音频失败，请重试')
-  }
-  return false // 阻止 antd 自动上传
-}
-
-// 音频拖拽处理（a-upload-dragger 会自动处理，这里只处理粘贴）
-async function handleAudioDrop(e) {
-  // a-upload-dragger 会自动处理拖拽，这里不需要额外处理
-  console.log('Audio drop event:', e)
-}
-
-// 音频粘贴处理
-function handleAudioPaste(e) {
-  const items = Array.from(e.clipboardData.items)
-  const audioItem = items.find(item => item.type.startsWith('audio/'))
-  
-  if (audioItem) {
-    const file = audioItem.getAsFile()
-    if (file) {
-      onUploadBefore(file)
+    const params = {
+      pageNo: pagination.current,
+      pageSize: pagination.pageSize,
+      name: searchParams.name || undefined
     }
-  }
-}
-
-function resetForm() {
-  Object.assign(form, {
-    id: '',
-    name: '',
-    language: 'zh-CN',
-    gender: 'female',
-    referenceAudio: '',
-    sampleText: '今天天气很好，我们一起去公园散步吧。',
-    enhancement: 50,
-    noiseReduction: true,
-    note: '',
-    originalText: '',
-  })
-}
-
-function validate() {
-  if (!form.referenceAudio) {
-    message.warning('请先上传或粘贴一段参考音频')
-    return false
-  }
-  return true
-}
-
-async function transcribeOriginal() {
-  const url = (form.referenceAudio || '').trim()
-  if (!url) { message.warning('请先提供参考音频'); return }
-  if (!/^https?:\/\//i.test(url)) {
-    message.info('当前仅支持网络音频链接一键转写，请粘贴 http(s) 链接')
-    return
-  }
-  try {
-    const list = await getVoiceText([{ audio_url: url }])
-    const text = Array.isArray(list) && list[0]?.value ? list[0].value : ''
-    if (text) {
-      form.originalText = text
-      message.success('已获取原语音文本')
+    const res = await VoiceService.getPage(params)
+    if (res.code === 0) {
+      voiceList.value = res.data.list || []
+      pagination.total = res.data.total || 0
     } else {
-      message.warning('未获取到可用文本，请稍后重试')
+      message.error(res.msg || '加载失败')
     }
-  } catch (e) {
-    console.error(e)
-    message.error('转写失败，请稍后重试')
+  } catch (error) {
+    console.error('加载配音列表失败:', error)
+    message.error('加载失败，请稍后重试')
+  } finally {
+    loading.value = false
   }
 }
 
-async function saveProfile() {
-  if (!validate()) return
-  if (!form.id) {
-    const created = await store.add({ ...form, id: '' })
-    Object.assign(form, { ...created })
-    message.success('已保存到本地')
-  } else {
-    const updated = await store.update({ ...form })
-    Object.assign(form, { ...updated })
-    message.success('已更新')
+// ========== 搜索和分页 ==========
+const handleSearch = () => {
+  pagination.current = 1
+  loadVoiceList()
+}
+
+const handleReset = () => {
+  searchParams.name = ''
+  pagination.current = 1
+  loadVoiceList()
+}
+
+const handleTableChange = (pag) => {
+  pagination.current = pag.current
+  pagination.pageSize = pag.pageSize
+  loadVoiceList()
+}
+
+// ========== CRUD 操作 ==========
+const handleCreate = () => {
+  formMode.value = 'create'
+  resetForm()
+  modalVisible.value = true
+}
+
+const handleEdit = async (record) => {
+  formMode.value = 'edit'
+  try {
+    const res = await VoiceService.get(record.id)
+    if (res.code === 0 && res.data) {
+      fillFormData(res.data)
+    } else {
+      fillFormData(record) // 获取失败时使用列表数据
+    }
+  } catch (error) {
+    console.error('获取配音详情失败:', error)
+    fillFormData(record) // 异常时使用列表数据
   }
+  modalVisible.value = true
 }
 
-function openSaveAs() {
-  if (!validate()) return
-  saveAsName.value = form.name ? `${form.name}-副本` : ''
-  isSavingAs.value = true
-}
-
-async function confirmSaveAs() {
-  const name = (saveAsName.value || '').trim()
-  if (!name) {
-    message.warning('请输入名称')
-    return
-  }
-  const created = await store.duplicate({ ...form }, name)
-  Object.assign(form, { ...created })
-  isSavingAs.value = false
-  message.success('已另存为')
-}
-
-function selectProfile(p) {
-  if (!p) return
-  store.select(p.id)
-  Object.assign(form, { ...p })
-}
-
-function requestDelete(p) {
+const handleDelete = (record) => {
   Modal.confirm({
-    title: '删除克隆声音',
-    content: `确定删除「${p.name || '未命名'}」吗？此操作不可恢复。`,
+    title: '确认删除',
+    content: `确定要删除配音「${record.name}」吗？此操作不可恢复。`,
     okText: '删除',
     okButtonProps: { danger: true },
     cancelText: '取消',
     onOk: async () => {
-      await store.remove(p.id)
-      if (!store.activeProfile) { resetForm() } else { Object.assign(form, { ...store.activeProfile }) }
-      message.success('已删除')
+      try {
+        const res = await VoiceService.delete(record.id)
+        if (res.code === 0) {
+          message.success('删除成功')
+          loadVoiceList()
+        } else {
+          message.error(res.msg || '删除失败')
+        }
+      } catch (error) {
+        console.error('删除失败:', error)
+        message.error('删除失败，请稍后重试')
+      }
     }
   })
 }
 
-onMounted(loadProfiles)
-</script>
-
-<template>
-  <div class="vc-page">
-    <div class="vc-grid">
-      <!-- 左侧：语音克隆表单 -->
-      <section class="vc-left">
-        <div class="vc-header">
-          <div class="vc-title">云端语音克隆</div>
-          <div class="vc-subtitle">上传参考音频，创建专属语音模型</div>
-        </div>
-        
-        <div class="vc-steps">
-          <div class="step-item" :class="{ active: form.referenceAudio }">
-            <div class="step-number">1</div>
-            <div class="step-label">上传音频</div>
-          </div>
-          <div class="step-divider"></div>
-          <div class="step-item" :class="{ active: form.referenceAudio && form.sampleText }">
-            <div class="step-number">2</div>
-            <div class="step-label">配置参数</div>
-          </div>
-          <div class="step-divider"></div>
-          <div class="step-item" :class="{ active: form.id }">
-            <div class="step-number">3</div>
-            <div class="step-label">保存模型</div>
-          </div>
-        </div>
-
-        <a-form layout="vertical" class="vc-form">
-          <!-- 基本信息 -->
-          <div class="form-section">
-            <div class="section-title">基本信息</div>
-            
-            <a-form-item label="模型名称" class="form-item">
-              <template #label>
-                <span class="form-label">模型名称</span>
-                <span class="form-label-hint">（可选）</span>
-              </template>
-              <a-input 
-                v-model:value="form.name" 
-                placeholder="为你的语音模型起个名字，如：温柔女声-普通话" 
-                allow-clear 
-                size="large"
-                class="form-input"
-              />
-            </a-form-item>
-
-            <div class="form-row">
-              <a-form-item label="语言" class="form-item">
-                <template #label>
-                  <span class="form-label">语言</span>
-                </template>
-                <a-select 
-                  v-model:value="form.language" 
-                  size="large"
-                  class="form-select"
-                  :options="[
-                    { value: 'zh-CN', label: '简体中文' },
-                    { value: 'zh-TW', label: '繁體中文' },
-                    { value: 'en-US', label: 'English' }
-                  ]" 
-                />
-              </a-form-item>
-
-              <a-form-item label="音色类型" class="form-item">
-                <template #label>
-                  <span class="form-label">音色类型</span>
-                </template>
-                <a-radio-group v-model:value="form.gender" size="large" class="form-radio-group">
-                  <a-radio-button value="female">女声</a-radio-button>
-                  <a-radio-button value="male">男声</a-radio-button>
-                </a-radio-group>
-              </a-form-item>
-            </div>
-          </div>
-
-          <!-- 参考音频 -->
-          <div class="form-section">
-            <div class="section-title">参考音频</div>
-            <div class="section-hint">上传清晰的参考音频，建议时长 10-60 秒</div>
-            
-            <a-form-item class="form-item">
-              <a-upload-dragger
-                name="file"
-                accept="audio/*"
-                action=""
-                :before-upload="onUploadBefore"
-                :show-upload-list="false"
-                @drop="handleAudioDrop"
-                @paste="handleAudioPaste"
-              >
-                <p class="ant-upload-drag-icon">
-                  <InboxOutlined />
-                </p>
-                <p class="ant-upload-text">点击或拖拽音频文件到此处上传</p>
-                <p class="ant-upload-hint">
-                  支持点击、拖拽或粘贴上传
-                  <br />
-                  支持格式：MP3、WAV、M4A、FLAC、AAC、OGG
-                  <br />
-                  文件大小 ≤ 50MB，时长 10-60 秒
-                </p>
-              </a-upload-dragger>
-              
-              <div class="audio-url-input">
-                <a-input 
-                  v-model:value="form.referenceAudio" 
-                  placeholder="或粘贴音频外链 URL" 
-                  size="large"
-                  class="form-input"
-                >
-                  <template #prefix>
-                    <svg viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" width="16" height="16">
-                      <path d="M10 13a5 5 0 0 0 7.54.54l3-3a5 5 0 0 0-7.07-7.07l-1.72 1.71" />
-                      <path d="M14 11a5 5 0 0 0-7.54-.54l-3 3a5 5 0 0 0 7.07 7.07l1.71-1.71" />
-                    </svg>
-                  </template>
-                </a-input>
-              </div>
-              
-              <div v-if="form.referenceAudio" class="audio-preview-card">
-                <div class="preview-header">
-                  <span class="preview-label">音频预览</span>
-                  <a-button type="link" danger @click="form.referenceAudio = ''" size="small">
-                    清除
-                  </a-button>
-                </div>
-                <audio :src="form.referenceAudio" controls preload="metadata" class="audio-player" />
-              </div>
-            </a-form-item>
-          </div>
-
-          <!-- 文本配置 -->
-          <div class="form-section">
-            <div class="section-title">文本配置</div>
-            
-            <a-form-item label="示例文本" class="form-item">
-              <template #label>
-                <span class="form-label">示例文本</span>
-                <span class="form-label-hint">（可选，可提升克隆效果）</span>
-              </template>
-              <a-textarea 
-                v-model:value="form.sampleText" 
-                :rows="3" 
-                placeholder="输入一段示例文本，建议与参考音频内容相关" 
-                class="form-textarea"
-              />
-            </a-form-item>
-
-            <a-form-item label="原语音文本" class="form-item">
-              <template #label>
-                <div class="label-with-action">
-                  <span class="form-label">原语音文本</span>
-                  <a-button 
-                    type="link" 
-                    size="small" 
-                    @click="transcribeOriginal"
-                    :disabled="!form.referenceAudio || !/^https?:\/\//i.test(form.referenceAudio)"
-                    class="action-btn"
-                  >
-                    一键转写
-                  </a-button>
-                </div>
-              </template>
-              <a-textarea 
-                v-model:value="form.originalText" 
-                :rows="4" 
-                placeholder="填写参考音频中的发音文本，或点击一键转写自动获取" 
-                class="form-textarea"
-              />
-              <div class="form-hint">仅支持网络链接的一键转写功能</div>
-            </a-form-item>
-          </div>
-
-          <!-- 高级设置 -->
-          <div class="form-section">
-            <div class="section-title">高级设置</div>
-            
-            <div class="form-row">
-              <a-form-item label="去噪优化" class="form-item">
-                <template #label>
-                  <span class="form-label">去噪优化</span>
-                </template>
-                <a-switch v-model:checked="form.noiseReduction" />
-                <div class="form-hint">自动去除背景噪音</div>
-              </a-form-item>
-
-              <a-form-item label="增强强度" class="form-item">
-                <template #label>
-                  <span class="form-label">增强强度</span>
-                  <span class="form-label-value">{{ form.enhancement }}%</span>
-                </template>
-                <a-slider 
-                  v-model:value="form.enhancement" 
-                  :min="0" 
-                  :max="100" 
-                  :tooltip-formatter="(val) => `${val}%`"
-                  class="form-slider"
-                />
-                <div class="form-hint">调整音频清晰度和去噪强度</div>
-              </a-form-item>
-            </div>
-
-            <a-form-item label="备注" class="form-item">
-              <template #label>
-                <span class="form-label">备注</span>
-                <span class="form-label-hint">（可选）</span>
-              </template>
-              <a-textarea 
-                v-model:value="form.note" 
-                :rows="2" 
-                placeholder="添加备注信息，便于后续管理" 
-                class="form-textarea"
-              />
-            </a-form-item>
-          </div>
-
-          <!-- 操作按钮 -->
-          <div class="form-actions">
-            <a-space size="middle">
-              <a-button 
-                type="primary" 
-                size="large"
-                @click="saveProfile"
-                :disabled="!form.referenceAudio"
-                class="action-btn-primary"
-              >
-                {{ form.id ? '更新模型' : '保存模型' }}
-              </a-button>
-              <a-button 
-                size="large"
-                @click="openSaveAs"
-                :disabled="!form.referenceAudio"
-              >
-                另存为
-              </a-button>
-              <a-button 
-                size="large"
-                @click="resetForm"
-              >
-                重置
-              </a-button>
-            </a-space>
-            <div class="form-hint" style="margin-top: 12px;">
-              保存后可在右侧管理你的语音模型
-            </div>
-          </div>
-        </a-form>
-      </section>
-
-      <!-- 右侧：已保存的克隆声音列表 -->
-      <section class="vc-right">
-        <div class="vc-header">
-          <div class="vc-title">我的语音模型</div>
-          <div class="vc-subtitle">共 {{ profiles.length }} 个模型</div>
-        </div>
-        
-        <div v-if="!profiles.length" class="vc-empty">
-          <div class="empty-icon">🎤</div>
-          <div class="empty-text">暂无语音模型</div>
-          <div class="empty-hint">创建并保存后，模型将显示在这里</div>
-        </div>
-        
-        <ul v-else class="vc-list">
-          <li 
-            v-for="p in profiles" 
-            :key="p.id" 
-            class="vc-item" 
-            :class="{ active: p.id === activeId }" 
-            @click="selectProfile(p)"
-          >
-            <div class="vc-item-main">
-              <div class="vc-item-header">
-                <div class="vc-item-name">{{ p.name || '未命名模型' }}</div>
-                <div class="vc-item-badges">
-                  <span class="badge">{{ p.language === 'zh-CN' ? '中文' : p.language === 'zh-TW' ? '繁體' : 'English' }}</span>
-                  <span class="badge">{{ p.gender === 'female' ? '女声' : '男声' }}</span>
-                </div>
-              </div>
-              <div class="vc-item-meta">
-                <span class="meta-item" v-if="p.note">{{ p.note }}</span>
-                <span class="meta-item" v-else>无备注</span>
-              </div>
-            </div>
-            <button 
-              class="vc-item-del" 
-              @click.stop="requestDelete(p)" 
-              aria-label="删除"
-              title="删除"
-            >
-              <svg viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" width="14" height="14">
-                <polyline points="3 6 5 6 21 6" />
-                <path d="M19 6v14a2 2 0 0 1-2 2H7a2 2 0 0 1-2-2V6m3 0V4a2 2 0 0 1 2-2h4a2 2 0 0 1 2 2v2" />
-              </svg>
-            </button>
-          </li>
-        </ul>
-      </section>
-    </div>
-
-    <a-modal 
-      v-model:open="isSavingAs" 
-      title="另存为" 
-      :maskClosable="false" 
-      @ok="confirmSaveAs" 
-      @cancel="() => (isSavingAs = false)"
-    >
-      <a-input 
-        v-model:value="saveAsName" 
-        placeholder="输入模型名称，如：小红-普通话-温柔女声" 
-        size="large"
-      />
-    </a-modal>
-  </div>
-</template>
-
-<style scoped>
-.vc-page { 
-  color: var(--color-text);
-  padding: 16px;
-}
-
-.vc-grid {
-  display: grid;
-  grid-template-columns: 1fr;
-  gap: 16px;
-  margin: 0 auto;
-}
-
-@media (min-width: 1024px) {
-  .vc-grid { 
-    grid-template-columns: 1fr 1fr; 
+// ========== 语音识别 ==========
+const handleTranscribe = async (record) => {
+  transcribingId.value = record.id
+  try {
+    const res = await VoiceService.transcribe(record.id)
+    if (res.code === 0) {
+      message.success('识别任务已提交，正在识别中...')
+      startPollingTranscription(record.id)
+    } else {
+      message.error(res.msg || '识别失败')
+      transcribingId.value = null
+    }
+  } catch (error) {
+    console.error('识别失败:', error)
+    message.error('识别失败，请稍后重试')
+    transcribingId.value = null
   }
 }
 
-.vc-left, .vc-right {
-  background: var(--color-surface);
-  border: 1px solid var(--color-border);
-  border-radius: var(--radius-card);
-  box-shadow: var(--shadow-inset-card);
-  padding: 20px;
+const stopPolling = () => {
+  if (pollingTimer) {
+    clearInterval(pollingTimer)
+    pollingTimer = null
+  }
+  transcribingId.value = null
 }
 
-/* 头部 */
-.vc-header {
-  margin-bottom: 20px;
-  padding-bottom: 16px;
-  border-bottom: 1px solid var(--color-border);
+const startPollingTranscription = (voiceId) => {
+  stopPolling()
+  
+  let pollCount = 0
+  pollingTimer = setInterval(async () => {
+    pollCount++
+    
+    try {
+      const res = await VoiceService.get(voiceId)
+      if (res.code === 0 && res.data?.transcription) {
+        stopPolling()
+        message.success('识别完成')
+        loadVoiceList()
+        return
+      }
+      
+      if (pollCount >= POLLING_CONFIG.maxCount) {
+        stopPolling()
+        message.warning('识别超时，请稍后手动刷新查看结果')
+        loadVoiceList()
+      }
+    } catch (error) {
+      console.error('轮询识别结果失败:', error)
+      if (pollCount >= POLLING_CONFIG.maxCount) {
+        stopPolling()
+      }
+    }
+  }, POLLING_CONFIG.interval)
 }
 
-.vc-title { 
-  font-size: 18px; 
-  font-weight: 600;
-  color: var(--color-text); 
-  margin-bottom: 4px; 
+// ========== 音频播放 ==========
+const handlePlayAudio = (record) => {
+  if (record.fileUrl && audioPlayer.value) {
+    audioPlayer.value.src = record.fileUrl
+    audioPlayer.value.play()
+  } else {
+    message.warning('音频文件不存在')
+  }
 }
 
-.vc-subtitle {
-  font-size: 13px;
-  color: var(--color-text-secondary);
-  font-weight: 400;
+// ========== 文件上传 ==========
+const handleBeforeUpload = (file) => {
+  // 检查文件大小（100MB）
+  const MAX_FILE_SIZE = 100 * 1024 * 1024
+  if (file.size > MAX_FILE_SIZE) {
+    message.error('文件大小不能超过 100MB')
+    return false
+  }
+
+  // 检查文件类型
+  const validTypes = ['audio/mpeg', 'audio/wav', 'audio/wave', 'audio/x-wav', 'audio/aac', 'audio/mp4', 'audio/flac', 'audio/ogg']
+  const validExtensions = ['.mp3', '.wav', '.aac', '.m4a', '.flac', '.ogg']
+  const fileName = file.name.toLowerCase()
+  const fileType = file.type.toLowerCase()
+  
+  const isValidType = validTypes.some(type => fileType.includes(type)) || 
+                      validExtensions.some(ext => fileName.endsWith(ext))
+  
+  if (!isValidType) {
+    message.error('请上传音频文件（MP3、WAV、AAC、M4A、FLAC、OGG）')
+    return false
+  }
+
+  return true // 允许添加到文件列表
 }
 
-/* 步骤指示器 */
-.vc-steps {
+const handleCustomUpload = async (options) => {
+  const { file, onSuccess, onError } = options
+  
+  uploading.value = true
+  
+  try {
+    const res = await MaterialService.uploadFile(file, 'voice', null)
+    
+    if (res.code === 0) {
+      formData.fileId = res.data
+      message.success('文件上传成功')
+      
+      // 使用 nextTick 确保 DOM 更新完成后再调用回调
+      await nextTick()
+      
+      // 安全调用 onSuccess
+      if (onSuccess && typeof onSuccess === 'function') {
+        try {
+          onSuccess(res, file)
+        } catch (err) {
+          console.warn('onSuccess 回调执行失败:', err)
+        }
+      }
+    } else {
+      const errorMsg = res.msg || '上传失败'
+      message.error(errorMsg)
+      
+      // 安全调用 onError
+      if (onError && typeof onError === 'function') {
+        try {
+          onError(new Error(errorMsg))
+        } catch (err) {
+          console.warn('onError 回调执行失败:', err)
+        }
+      }
+    }
+  } catch (error) {
+    console.error('上传失败:', error)
+    const errorMsg = error?.message || '上传失败，请稍后重试'
+    message.error(errorMsg)
+    
+    // 安全调用 onError
+    if (onError && typeof onError === 'function') {
+      try {
+        onError(error)
+      } catch (err) {
+        console.warn('onError 回调执行失败:', err)
+      }
+    }
+  } finally {
+    uploading.value = false
+  }
+}
+
+const handleFileListChange = (info) => {
+  // 处理文件列表变化，避免直接修改导致 DOM 错误
+  const { fileList: newFileList } = info
+  
+  // 只更新文件列表，不直接修改文件项的状态
+  // 让组件自己管理状态
+  if (newFileList) {
+    fileList.value = newFileList.filter(item => item.status !== 'removed')
+  }
+}
+
+const handleRemoveFile = () => {
+  formData.fileId = null
+  fileList.value = []
+}
+
+// ========== 表单操作 ==========
+const handleSubmit = async () => {
+  try {
+    await formRef.value.validate()
+    submitting.value = true
+
+    const params = isCreateMode.value
+      ? {
+          name: formData.name,
+          fileId: formData.fileId,
+          autoTranscribe: formData.autoTranscribe,
+          language: formData.language,
+          gender: formData.gender,
+          note: formData.note
+        }
+      : {
+          id: formData.id,
+          name: formData.name,
+          language: formData.language,
+          gender: formData.gender,
+          note: formData.note,
+          transcription: formData.transcription
+        }
+
+    const res = isCreateMode.value
+      ? await VoiceService.create(params)
+      : await VoiceService.update(params)
+
+    if (res.code === 0) {
+      message.success(isCreateMode.value ? '创建成功' : '更新成功')
+      modalVisible.value = false
+      
+      // 如果开启了自动识别，开始轮询识别结果
+      if (isCreateMode.value && formData.autoTranscribe && res.data) {
+        const voiceId = res.data
+        message.info('自动识别已启动，正在识别中...')
+        startPollingTranscription(voiceId)
+      }
+      
+      loadVoiceList()
+    } else {
+      message.error(res.msg || '操作失败')
+    }
+  } catch (error) {
+    if (error?.errorFields) {
+      // 表单验证失败，不显示错误
+      return
+    }
+    console.error('提交失败:', error)
+    message.error('操作失败，请稍后重试')
+  } finally {
+    submitting.value = false
+  }
+}
+
+const handleCancel = () => {
+  modalVisible.value = false
+  resetForm()
+}
+
+const resetForm = () => {
+  Object.assign(formData, { ...DEFAULT_FORM_DATA })
+  fileList.value = []
+  formRef.value?.resetFields()
+}
+
+// ========== 生命周期 ==========
+onMounted(() => {
+  loadVoiceList()
+})
+
+onUnmounted(() => {
+  stopPolling()
+})
+</script>
+
+<style scoped>
+.voice-copy-page {
+  padding: 24px;
+  background: var(--color-bg);
+}
+
+.page-header {
   display: flex;
+  justify-content: space-between;
   align-items: center;
-  gap: 8px;
   margin-bottom: 24px;
-  padding: 12px;
-  background: rgba(0, 0, 0, 0.2);
-  border-radius: var(--radius-card);
-  border: 1px solid var(--color-border);
 }
 
-.step-item {
-  display: flex;
-  flex-direction: column;
+.page-header .ant-btn {
+  display: inline-flex;
   align-items: center;
-  gap: 6px;
-  flex: 1;
-  position: relative;
 }
 
-.step-number {
-  width: 32px;
-  height: 32px;
-  border-radius: 50%;
-  background: rgba(59, 130, 246, 0.1);
-  border: 2px solid rgba(59, 130, 246, 0.3);
-  color: var(--color-text-secondary);
-  display: flex;
-  align-items: center;
-  justify-content: center;
-  font-size: 14px;
-  font-weight: 600;
-  transition: all 0.3s ease;
-}
-
-.step-item.active .step-number {
-  background: var(--color-primary);
-  border-color: var(--color-primary);
-  color: white;
-  box-shadow: 0 2px 8px rgba(59, 130, 246, 0.3);
-}
-
-.step-label {
-  font-size: 12px;
-  color: var(--color-text-secondary);
-  font-weight: 500;
-}
-
-.step-item.active .step-label {
-  color: var(--color-text);
-}
-
-.step-divider {
-  width: 1px;
-  height: 20px;
-  background: var(--color-border);
-  margin: 0 4px;
-}
-
-/* 表单样式 */
-.vc-form {
-  display: flex;
-  flex-direction: column;
-  gap: 0;
-}
-
-.form-section {
-  margin-bottom: 24px;
-  padding-bottom: 24px;
-  border-bottom: 1px solid var(--color-border);
-}
-
-.form-section:last-of-type {
-  border-bottom: none;
-  margin-bottom: 0;
-  padding-bottom: 0;
-}
-
-.section-title {
-  font-size: 15px;
+.page-title {
+  font-size: 20px;
   font-weight: 600;
   color: var(--color-text);
-  margin-bottom: 8px;
+  margin: 0;
+  line-height: 1.5;
   display: flex;
   align-items: center;
-  gap: 8px;
 }
 
-.section-title::before {
-  content: '';
-  width: 3px;
-  height: 16px;
-  background: var(--color-primary);
-  border-radius: 2px;
-}
-
-.section-hint {
-  font-size: 12px;
-  color: var(--color-text-secondary);
+.search-bar {
   margin-bottom: 16px;
+  padding: 16px;
+  background: var(--color-surface);
+  border-radius: var(--radius-card);
+}
+
+.table-container {
+  background: var(--color-surface);
+  border-radius: var(--radius-card);
+  padding: 16px;
+}
+
+.voice-name {
+  font-weight: 500;
+  color: var(--color-text);
+}
+
+.transcription-text {
+  color: var(--color-text-secondary);
+  font-size: 13px;
   line-height: 1.5;
 }
 
-.form-item {
-  margin-bottom: 20px;
+.uploaded-file-info {
+  margin-top: 8px;
 }
 
-.form-item:last-child {
-  margin-bottom: 0;
-}
-
-.form-label {
-  font-size: 14px;
-  font-weight: 500;
-  color: var(--color-text);
-}
-
-.form-label-hint {
+.upload-hint {
   font-size: 12px;
   color: var(--color-text-secondary);
-  font-weight: 400;
-  margin-left: 4px;
-}
-
-.form-label-value {
-  font-size: 12px;
-  color: var(--color-primary);
-  font-weight: 600;
-  margin-left: 8px;
-}
-
-.label-with-action {
-  display: flex;
-  align-items: center;
-  justify-content: space-between;
-  width: 100%;
-}
-
-.action-btn {
-  padding: 0;
-  height: auto;
-  font-size: 12px;
-}
-
-.form-row {
-  display: grid;
-  grid-template-columns: 1fr 1fr;
-  gap: 16px;
-}
-
-.form-input,
-.form-textarea,
-.form-select {
-  border-radius: var(--radius-card);
-  border: 1px solid var(--color-border);
-  background: var(--color-bg);
-  color: var(--color-text);
-  transition: all 0.2s ease;
-}
-
-.form-input:focus,
-.form-textarea:focus,
-.form-select:focus {
-  border-color: var(--color-primary);
-  box-shadow: 0 0 0 2px rgba(59, 130, 246, 0.1);
-}
-
-.form-textarea {
-  resize: vertical;
-}
-
-.form-radio-group {
-  width: 100%;
-}
-
-.form-radio-group :deep(.ant-radio-button-wrapper) {
-  flex: 1;
-  text-align: center;
-  border-color: var(--color-border);
-  background: var(--color-bg);
-  color: var(--color-text-secondary);
-}
-
-.form-radio-group :deep(.ant-radio-button-wrapper-checked) {
-  background: var(--color-primary);
-  border-color: var(--color-primary);
-  color: white;
-}
-
-.form-slider {
-  margin: 8px 0;
+  margin-top: 8px;
+  line-height: 1.5;
 }
 
 .form-hint {
   font-size: 12px;
   color: var(--color-text-secondary);
-  margin-top: 6px;
-  line-height: 1.4;
-}
-
-/* 音频上传区域 */
-.audio-upload-zone {
-  width: 100%;
-  min-height: 200px;
-  border: 1.5px dashed rgba(59, 130, 246, 0.4);
-  border-radius: 8px;
-  padding: 40px 20px;
-  text-align: center;
-  cursor: pointer;
-  transition: all 0.3s cubic-bezier(0.4, 0, 0.2, 1);
-  background: rgba(0, 0, 0, 0.2);
-  position: relative;
-  overflow: hidden;
-  margin-bottom: 12px;
-}
-
-.audio-upload-zone.dragging {
-  border-color: var(--color-primary);
-  background: rgba(59, 130, 246, 0.12);
-  border-style: solid;
-  border-width: 2px;
-}
-
-.audio-upload-zone:hover {
-  border-color: var(--color-primary);
-  background: rgba(59, 130, 246, 0.08);
-  border-style: solid;
-}
-
-.upload-zone-content {
-  display: flex;
-  flex-direction: column;
-  align-items: center;
-  justify-content: center;
-  gap: 16px;
-  position: relative;
-  z-index: 1;
-}
-
-.upload-icon-wrapper {
-  width: 64px;
-  height: 64px;
-  display: flex;
-  align-items: center;
-  justify-content: center;
-  border-radius: 50%;
-  background: rgba(59, 130, 246, 0.1);
-  margin-bottom: 4px;
-  transition: all 0.3s ease;
-}
-
-.audio-upload-zone:hover .upload-icon-wrapper {
-  background: rgba(59, 130, 246, 0.15);
-  transform: scale(1.05);
-}
-
-.audio-upload-zone.dragging .upload-icon-wrapper {
-  background: rgba(59, 130, 246, 0.2);
-  transform: scale(1.1);
-}
-
-.upload-icon-svg {
-  width: 32px;
-  height: 32px;
-  color: var(--color-primary);
-  stroke-width: 1.5;
-}
-
-.upload-main-text {
-  font-size: 16px;
-  font-weight: 500;
-  color: var(--color-text);
   margin-top: 4px;
 }
-
-.upload-sub-text {
-  font-size: 14px;
-  color: var(--color-text-secondary);
-  font-weight: 400;
-}
-
-.format-tags {
-  display: flex;
-  flex-wrap: wrap;
-  gap: 6px;
-  justify-content: center;
-  margin-top: 4px;
-}
-
-.format-tag {
-  display: inline-block;
-  padding: 4px 10px;
-  font-size: 12px;
-  font-weight: 500;
-  color: var(--color-primary);
-  background: rgba(59, 130, 246, 0.1);
-  border: 1px solid rgba(59, 130, 246, 0.2);
-  border-radius: 4px;
-  line-height: 1.2;
-}
-
-.upload-limit-text {
-  font-size: 12px;
-  color: var(--color-text-secondary);
-  display: flex;
-  align-items: center;
-  gap: 8px;
-  flex-wrap: wrap;
-  justify-content: center;
-  margin-top: 4px;
-}
-
-.limit-item {
-  color: var(--color-text-secondary);
-}
-
-.limit-separator {
-  opacity: 0.4;
-  font-size: 10px;
-}
-
-.audio-url-input {
-  margin-top: 12px;
-}
-
-.audio-preview-card {
-  margin-top: 12px;
-  padding: 12px;
-  background: var(--color-bg);
-  border-radius: var(--radius-card);
-  border: 1px solid var(--color-border);
-}
-
-.preview-header {
-  display: flex;
-  align-items: center;
-  justify-content: space-between;
-  margin-bottom: 8px;
-}
-
-.preview-label {
-  font-size: 12px;
-  font-weight: 600;
-  color: var(--color-text);
-}
-
-.audio-player {
-  width: 100%;
-  margin-top: 4px;
-}
-
-/* 操作按钮 */
-.form-actions {
-  margin-top: 24px;
-  padding-top: 20px;
-  border-top: 1px solid var(--color-border);
-}
-
-.action-btn-primary {
-  min-width: 120px;
-}
-
-/* 右侧列表 */
-.vc-empty {
-  padding: 60px 20px;
-  text-align: center;
-  color: var(--color-text-secondary);
-}
-
-.empty-icon {
-  font-size: 48px;
-  margin-bottom: 12px;
-  opacity: 0.5;
-}
-
-.empty-text {
-  font-size: 16px;
-  font-weight: 500;
-  color: var(--color-text);
-  margin-bottom: 8px;
-}
-
-.empty-hint {
-  font-size: 12px;
-  color: var(--color-text-secondary);
-}
-
-.vc-list {
-  list-style: none;
-  padding: 0;
-  margin: 0;
-  display: flex;
-  flex-direction: column;
-  gap: 12px;
-}
-
-.vc-item {
-  display: flex;
-  align-items: center;
-  justify-content: space-between;
-  background: var(--color-bg);
-  border: 1px solid var(--color-border);
-  border-radius: var(--radius-card);
-  padding: 14px 16px;
-  cursor: pointer;
-  transition: all 0.2s ease;
-  position: relative;
-}
-
-.vc-item:hover {
-  background: rgba(59, 130, 246, 0.05);
-  border-color: rgba(59, 130, 246, 0.3);
-  transform: translateY(-1px);
-  box-shadow: 0 2px 8px rgba(59, 130, 246, 0.1);
-}
-
-.vc-item.active {
-  border-color: var(--color-primary);
-  background: rgba(59, 130, 246, 0.08);
-  box-shadow: 0 0 0 2px rgba(59, 130, 246, 0.1);
-}
-
-.vc-item-main {
-  flex: 1;
-  display: flex;
-  flex-direction: column;
-  gap: 8px;
-}
-
-.vc-item-header {
-  display: flex;
-  align-items: center;
-  justify-content: space-between;
-  gap: 12px;
-}
-
-.vc-item-name {
-  font-size: 15px;
-  color: var(--color-text);
-  font-weight: 600;
-  flex: 1;
-}
-
-.vc-item-badges {
-  display: flex;
-  gap: 6px;
-  flex-shrink: 0;
-}
-
-.badge {
-  padding: 2px 8px;
-  font-size: 11px;
-  font-weight: 500;
-  color: var(--color-primary);
-  background: rgba(59, 130, 246, 0.1);
-  border: 1px solid rgba(59, 130, 246, 0.2);
-  border-radius: 12px;
-  line-height: 1.4;
-}
-
-.vc-item-meta {
-  display: flex;
-  align-items: center;
-  gap: 8px;
-}
-
-.meta-item {
-  font-size: 12px;
-  color: var(--color-text-secondary);
-}
-
-.vc-item-del {
-  visibility: hidden;
-  width: 28px;
-  height: 28px;
-  border-radius: 6px;
-  background: rgba(239, 68, 68, 0.1);
-  color: #ef4444;
-  border: 1px solid rgba(239, 68, 68, 0.2);
-  cursor: pointer;
-  display: flex;
-  align-items: center;
-  justify-content: center;
-  transition: all 0.2s ease;
-  flex-shrink: 0;
-}
-
-.vc-item:hover .vc-item-del {
-  visibility: visible;
-}
-
-.vc-item-del:hover {
-  background: rgba(239, 68, 68, 0.2);
-  border-color: rgba(239, 68, 68, 0.4);
-  transform: scale(1.1);
-}
 </style>
diff --git a/frontend/app/web-gold/src/views/material/MaterialList.vue b/frontend/app/web-gold/src/views/material/MaterialList.vue
index 6c5582b285..0397e61001 100644
--- a/frontend/app/web-gold/src/views/material/MaterialList.vue
+++ b/frontend/app/web-gold/src/views/material/MaterialList.vue
@@ -9,6 +9,14 @@
           </template>
           上传素材
         </a-button>
+        <a-button
+          type="primary"
+          ghost
+          :disabled="selectedFileIds.length === 0"
+          @click="handleOpenMixModal"
+        >
+          素材混剪
+        </a-button>
         <a-button
           v-if="selectedFileIds.length > 0"
           type="primary"
@@ -142,11 +150,47 @@
       @confirm="handleConfirmUpload"
       @cancel="handleUploadCancel"
     />
+    <a-modal
+      v-model:open="mixModalVisible"
+      title="素材混剪"
+      centered
+      :confirm-loading="mixing"
+      ok-text="提交混剪"
+      cancel-text="取消"
+      @ok="handleMixConfirm"
+      @cancel="handleMixCancel"
+    >
+      <div class="mix-modal__summary">
+        <p>选中素材：{{ selectedFiles.length }} 个</p>
+        <p>视频素材：{{ selectedVideoUrls.length }} 个</p>
+        <p>背景音乐：{{ selectedAudioUrls.length }} 个</p>
+      </div>
+      <a-form layout="vertical">
+        <a-form-item label="视频标题" required>
+          <a-input v-model:value="mixForm.title" placeholder="请输入生成视频标题" />
+        </a-form-item>
+        <a-form-item label="文案内容" required>
+          <a-textarea
+            v-model:value="mixForm.text"
+            placeholder="请输入文案（每句话换行以便自动拆分）"
+            :rows="4"
+          />
+        </a-form-item>
+        <a-form-item label="生成成片数量" required>
+          <a-input-number
+            v-model:value="mixForm.produceCount"
+            :min="1"
+            :max="10"
+            style="width: 100%"
+          />
+        </a-form-item>
+      </a-form>
+    </a-modal>
   </div>
 </template>
 
 <script setup>
-import { ref, reactive, onMounted } from 'vue'
+import { ref, reactive, computed, onMounted } from 'vue'
 import { message, Modal } from 'ant-design-vue'
 import {
   UploadOutlined,
@@ -154,6 +198,7 @@ import {
   FileOutlined
 } from '@ant-design/icons-vue'
 import { MaterialService } from '@/api/material'
+import { MixService } from '@/api/mix'
 import MaterialUploadModal from '@/components/material/MaterialUploadModal.vue'
 import { formatFileSize, formatDate } from '@/utils/file'
 
@@ -163,6 +208,8 @@ const fileList = ref([])
 const selectedFileIds = ref([])
 const uploadModalVisible = ref(false)
 const uploading = ref(false)
+const mixModalVisible = ref(false)
+const mixing = ref(false)
 
 // 筛选条件
 const filters = reactive({
@@ -370,6 +417,111 @@ const handleImageError = (e) => {
   img.style.display = 'none'
 }
 
+const selectedFiles = computed(() =>
+  fileList.value.filter((file) => selectedFileIds.value.includes(file.id))
+)
+
+const isVideoFile = (file) => {
+  if (!file) return false
+  if (file.isVideo) return true
+  if (file.fileCategory === 'video') return true
+  if (typeof file.fileType === 'string' && file.fileType.startsWith('video')) return true
+  return false
+}
+
+const isAudioFile = (file) => {
+  if (!file) return false
+  if (file.fileCategory === 'audio') return true
+  if (file.fileType === 'audio') return true
+  if (typeof file.fileType === 'string' && file.fileType.startsWith('audio')) return true
+  return false
+}
+
+const selectedVideoUrls = computed(() =>
+  selectedFiles.value.map((file) => (isVideoFile(file) ? file?.fileUrl || file?.previewUrl : null)).filter(Boolean)
+)
+
+const selectedAudioUrls = computed(() =>
+  selectedFiles.value.map((file) => (isAudioFile(file) ? file?.fileUrl || file?.previewUrl : null)).filter(Boolean)
+)
+
+const mixForm = reactive({
+  title: '',
+  text: '',
+  produceCount: 1
+})
+
+const resetMixForm = () => {
+  mixForm.title = ''
+  mixForm.text = ''
+  mixForm.produceCount = 1
+}
+
+const handleOpenMixModal = () => {
+  if (selectedFileIds.value.length === 0) {
+    message.warning('请先选择至少一个素材')
+    return
+  }
+  if (selectedVideoUrls.value.length === 0) {
+    message.warning('请至少选择一个视频素材')
+    return
+  }
+  if (selectedAudioUrls.value.length === 0) {
+    message.warning('请至少选择一个背景音乐素材')
+    return
+  }
+  mixModalVisible.value = true
+}
+
+const handleMixCancel = () => {
+  mixModalVisible.value = false
+}
+
+const handleMixConfirm = async () => {
+  const title = mixForm.title.trim()
+  const text = mixForm.text.trim()
+  if (!title) {
+    message.warning('请输入视频标题')
+    return
+  }
+  if (!text) {
+    message.warning('请输入文案内容')
+    return
+  }
+  const produceCount = Math.max(1, Math.min(10, Number(mixForm.produceCount) || 1))
+  if (selectedVideoUrls.value.length === 0) {
+    message.warning('请至少选择一个视频素材')
+    return
+  }
+  if (selectedAudioUrls.value.length === 0) {
+    message.warning('请至少选择一个背景音乐素材')
+    return
+  }
+  mixing.value = true
+  try {
+    const { data } = await MixService.batchProduceAlignment({
+      title,
+      text,
+      videoUrls: selectedVideoUrls.value,
+      bgMusicUrls: selectedAudioUrls.value,
+      produceCount
+    })
+    const jobIds = Array.isArray(data) ? data : []
+    message.success(
+      jobIds.length > 0
+        ? `混剪任务提交成功，JobId：${jobIds.join(', ')}`
+        : '混剪任务提交成功'
+    )
+    mixModalVisible.value = false
+    resetMixForm()
+  } catch (error) {
+    console.error('混剪失败:', error)
+    message.error(error?.message || '混剪任务提交失败，请重试')
+  } finally {
+    mixing.value = false
+  }
+}
+
 // 初始化
 onMounted(() => {
   loadFileList()
@@ -524,5 +676,20 @@ onMounted(() => {
   color: var(--color-text-3);
 }
 
+.mix-modal__summary {
+  margin-bottom: 16px;
+  padding: 12px;
+  background: var(--color-bg-2);
+  border: 1px dashed var(--color-border);
+  border-radius: var(--radius-card);
+  font-size: 13px;
+  color: var(--color-text-2);
+}
+
+.mix-modal__summary p {
+  margin: 0;
+  line-height: 1.6;
+}
+
 </style>
 
diff --git a/frontend/app/web-gold/src/views/mix/MixEditor.vue b/frontend/app/web-gold/src/views/mix/MixEditor.vue
deleted file mode 100644
index bd4f7957be..0000000000
--- a/frontend/app/web-gold/src/views/mix/MixEditor.vue
+++ /dev/null
@@ -1,22 +0,0 @@
-<script setup>
-</script>
-
-<template>
-  <div class="space-y-4">
-    <h2 class="text-xl font-bold">素材混剪</h2>
-    <div class="grid grid-cols-1 lg:grid-cols-3 gap-4">
-      <section class="bg-white p-4 rounded shadow lg:col-span-1">
-        <div class="text-gray-600 text-sm">文案拆解与镜头建议。</div>
-      </section>
-      <section class="bg-white p-4 rounded shadow lg:col-span-2">
-        <div class="text-gray-500">素材匹配与时间线，导出到剪映。</div>
-      </section>
-    </div>
-  </div>
-</template>
-
-<style scoped>
-</style>
-
-
-
diff --git a/yudao-module-infra/src/main/java/cn/iocoder/yudao/module/infra/framework/file/core/client/s3/S3FileClient.java b/yudao-module-infra/src/main/java/cn/iocoder/yudao/module/infra/framework/file/core/client/s3/S3FileClient.java
index 94ba6a3ebb..00d3511966 100644
--- a/yudao-module-infra/src/main/java/cn/iocoder/yudao/module/infra/framework/file/core/client/s3/S3FileClient.java
+++ b/yudao-module-infra/src/main/java/cn/iocoder/yudao/module/infra/framework/file/core/client/s3/S3FileClient.java
@@ -3,6 +3,7 @@ package cn.iocoder.yudao.module.infra.framework.file.core.client.s3;
 import cn.hutool.core.io.IoUtil;
 import cn.hutool.core.util.BooleanUtil;
 import cn.hutool.core.util.StrUtil;
+import cn.hutool.core.util.URLUtil;
 import cn.hutool.http.HttpUtil;
 import cn.iocoder.yudao.framework.common.util.http.HttpUtils;
 import cn.iocoder.yudao.module.infra.framework.file.core.client.AbstractFileClient;
@@ -23,6 +24,9 @@ import software.amazon.awssdk.services.s3.presigner.model.PutObjectPresignReques
 import java.net.URI;
 import java.net.URL;
 import java.time.Duration;
+import java.nio.charset.StandardCharsets;
+
+import org.springframework.web.util.UriUtils;
 
 /**
  * 基于 S3 协议的文件客户端，实现 MinIO、阿里云、腾讯云、七牛云、华为云等云服务
@@ -115,15 +119,17 @@ public class S3FileClient extends AbstractFileClient<S3FileClientConfig> {
         // 1. 将 url 转换为 path
         String path = StrUtil.removePrefix(url, config.getDomain() + "/");
         path = HttpUtils.removeUrlQuery(path);
+        String decodedPath = URLUtil.decode(path, StandardCharsets.UTF_8);
 
         // 2.1 情况一：公开访问：无需签名
         // 考虑到老版本的兼容，所以必须是 config.getEnablePublicAccess() 为 false 时，才进行签名
         if (!BooleanUtil.isFalse(config.getEnablePublicAccess())) {
-            return config.getDomain() + "/" + path;
+            String encodedPath = UriUtils.encodePath(decodedPath, StandardCharsets.UTF_8);
+            return config.getDomain() + "/" + encodedPath;
         }
 
         // 2.2 情况二：私有访问：生成 GET 预签名 URL
-        String finalPath = path;
+        String finalPath = decodedPath;
         Duration expiration = expirationSeconds != null ? Duration.ofSeconds(expirationSeconds) : EXPIRATION_DEFAULT;
         URL signedUrl = presigner.presignGetObject(GetObjectPresignRequest.builder()
                 .signatureDuration(expiration)
diff --git a/yudao-module-tik/DESIGN.md b/yudao-module-tik/DESIGN.md
deleted file mode 100644
index e2a3d84a12..0000000000
--- a/yudao-module-tik/DESIGN.md
+++ /dev/null
@@ -1,282 +0,0 @@
-# Tik 文件管理模块设计文档
-
-## 一、模块概述
-
-Tik 文件管理模块负责用户文件的上传、存储、管理和分组功能，支持多种文件类型（视频、图片、音频等）和分类管理。
-
-## 二、表结构设计
-
-### 2.1 核心表
-
-#### 1. `tik_user_file` - 用户文件表
-**作用**：存储用户上传的文件元数据
-
-**关键字段**：
-- `file_path` (varchar(1024)): **完整OSS路径**，格式：`{手机号MD5}/{租户ID}/{分类}/{日期}/{文件名}_{时间戳}.ext`
-- `file_url` (varchar(1024)): 文件访问URL（预签名URL或公开URL）
-- `oss_root_path` (varchar(256)): OSS根路径，用于快速定位用户文件目录
-- `file_category`: 文件分类（video/generate/audio/mix/voice）
-- `file_id`: 关联 `infra_file.id`（可选，用于关联系统文件表）
-
-**索引设计**：
-- `idx_user_id`: 用户ID索引
-- `idx_file_category`: 文件分类索引
-- `idx_user_tenant`: 用户+租户联合索引
-- `idx_create_time`: 创建时间索引
-
-#### 2. `tik_user_oss_init` - OSS初始化记录表
-**作用**：记录用户OSS目录初始化状态和路径信息
-
-**关键字段**：
-- `mobile_md5`: 手机号MD5值（用于生成OSS路径）
-- `oss_root_path`: OSS根路径
-- `video_path`, `generate_path`, `audio_path`, `mix_path`, `voice_path`: 各分类目录路径
-- `init_status`: 初始化状态（0-未初始化，1-已初始化）
-
-**设计要点**：
-- 懒加载策略：首次上传时自动初始化
-- 路径格式：`{手机号MD5}/{租户ID}/{分类}`
-
-#### 3. `tik_file_group` - 文件分组表
-**作用**：用户自定义文件分组（支持层级分组）
-
-**关键字段**：
-- `parent_id`: 父分组ID（0表示根分组）
-- `sort`: 排序字段
-
-#### 4. `tik_user_file_group` - 文件分组关联表
-**作用**：文件与分组的关联关系（支持一个文件属于多个分组）
-
-**设计要点**：
-- 多对多关系
-- 唯一索引：`uk_file_group` (file_id, group_id)
-
-#### 5. `tik_user_quota` - 用户配额表
-**作用**：管理用户存储配额和VIP等级
-
-**关键字段**：
-- `total_storage`: 总存储空间（字节）
-- `used_storage`: 已使用存储空间（字节）
-- `vip_level`: VIP等级
-
-## 三、架构设计
-
-### 3.1 分层架构
-
-```
-Controller 层 (AppTikUserFileController)
-    ↓
-Service 层 (TikUserFileService)
-    ↓
-Mapper 层 (TikUserFileMapper)
-    ↓
-DataObject 层 (TikUserFileDO)
-```
-
-### 3.2 核心服务
-
-#### 1. TikUserFileService - 文件管理服务
-**职责**：
-- 文件上传（带配额校验）
-- 文件查询（分页、筛选）
-- 文件删除（逻辑删除 + 物理删除）
-- 预签名URL生成
-
-**关键流程**：
-1. **上传流程**：
-   ```
-   校验文件分类 → 校验配额 → 获取OSS目录 → 生成完整路径 → 上传到OSS → 保存元数据 → 更新配额
-   ```
-
-2. **删除流程**：
-   ```
-   校验权限 → 物理删除OSS文件 → 逻辑删除记录 → 释放配额
-   ```
-
-#### 2. TikOssInitService - OSS初始化服务
-**职责**：
-- 初始化用户OSS目录结构
-- 获取OSS路径信息
-- 懒加载策略实现
-
-**设计要点**：
-- OSS目录是虚拟的，不需要显式创建
-- 首次上传时自动初始化
-- 路径格式：`{手机号MD5}/{租户ID}/{分类}`
-
-#### 3. TikFileGroupService - 文件分组服务
-**职责**：
-- 分组CRUD
-- 层级分组支持
-
-#### 4. TikUserQuotaService - 配额管理服务
-**职责**：
-- 配额校验
-- 配额更新
-- VIP等级管理
-
-## 四、路径设计
-
-### 4.1 OSS路径结构
-
-```
-{手机号MD5}/{租户ID}/{分类}/{日期}/{文件名}_{时间戳}.ext
-```
-
-**示例**：
-```
-abc123def45678901234567890123456/1/video/20250101/my_video_1234567890123.mp4
-```
-
-**路径组成部分**：
-1. **手机号MD5** (32字符): 用户唯一标识，保护隐私
-2. **租户ID**: 多租户隔离
-3. **分类** (video/generate/audio/mix/voice): 文件分类
-4. **日期** (yyyyMMdd): 按日期分目录，便于管理
-5. **文件名+时间戳**: 保证唯一性，避免覆盖
-
-### 4.2 路径存储策略
-
-- **file_path**: 存储完整OSS路径（用于物理删除）
-- **file_url**: 存储访问URL（用于前端展示）
-- **oss_root_path**: 存储根路径（用于快速定位）
-
-## 五、设计亮点
-
-### 5.1 优点
-
-1. **分层清晰**：Controller → Service → Mapper → DO，职责明确
-2. **配额管理**：上传前校验，删除后释放
-3. **多租户支持**：通过 tenant_id 隔离
-4. **懒加载策略**：OSS目录按需初始化
-5. **路径设计合理**：包含用户、租户、分类、日期等信息
-6. **分组功能**：支持多分组、层级分组
-
-### 5.2 需要改进的地方
-
-1. **物理删除OSS文件**：
-   - 当前只做了逻辑删除，OSS文件未删除
-   - 建议：删除时调用 FileService 或 FileClient 删除OSS文件
-   - 或者：定期清理已逻辑删除的文件
-
-2. **file_path 字段长度**：
-   - 当前：varchar(512)
-   - 建议：varchar(1024) 更安全
-
-3. **文件关联 infra_file 表**：
-   - `file_id` 字段存在但未充分利用
-   - 建议：上传时关联 infra_file 表，便于统一管理
-
-4. **预览图生成**：
-   - 视频封面和图片缩略图功能未实现
-   - 建议：异步生成预览图
-
-5. **批量操作优化**：
-   - 删除文件时逐个删除OSS文件，可能较慢
-   - 建议：批量删除或异步删除
-
-## 六、数据流
-
-### 6.1 上传流程
-
-```
-前端上传文件
-  ↓
-Controller 接收
-  ↓
-Service 校验（分类、配额）
-  ↓
-获取OSS目录（懒加载初始化）
-  ↓
-生成完整路径
-  ↓
-上传到OSS（FileApi）
-  ↓
-保存元数据到 tik_user_file
-  ↓
-更新配额（tik_user_quota）
-  ↓
-返回文件ID
-```
-
-### 6.2 查询流程
-
-```
-前端请求文件列表
-  ↓
-Controller 接收查询参数
-  ↓
-Service 查询数据库（分页、筛选）
-  ↓
-转换为VO（生成预览URL）
-  ↓
-返回分页结果
-```
-
-### 6.3 删除流程
-
-```
-前端请求删除
-  ↓
-Controller 接收文件ID列表
-  ↓
-Service 校验权限
-  ↓
-物理删除OSS文件（TODO）
-  ↓
-逻辑删除数据库记录
-  ↓
-释放配额
-  ↓
-返回成功
-```
-
-## 七、API设计
-
-### 7.1 文件管理API
-
-- `POST /api/tik/file/upload` - 上传文件
-- `GET /api/tik/file/page` - 分页查询
-- `DELETE /api/tik/file/delete-batch` - 批量删除
-- `GET /api/tik/file/video/play-url` - 获取视频播放URL
-- `GET /api/tik/file/audio/play-url` - 获取音频播放URL
-- `GET /api/tik/file/preview-url` - 获取预览URL
-
-### 7.2 分组管理API
-
-- `POST /api/tik/file/group/create` - 创建分组
-- `PUT /api/tik/file/group/update` - 更新分组
-- `DELETE /api/tik/file/group/delete` - 删除分组
-- `GET /api/tik/file/group/list` - 查询分组列表
-- `POST /api/tik/file/group/add-files` - 添加文件到分组
-- `POST /api/tik/file/group/remove-files` - 从分组移除文件
-
-## 八、总结
-
-### 8.1 表结构建议
-
-1. **必须修改**：
-   - `file_path` 字段长度：512 → 1024
-
-2. **可选优化**：
-   - 添加 `file_path` 索引（如果经常按路径查询）
-   - 添加 `file_id` 索引（如果关联 infra_file 表）
-
-### 8.2 功能完善建议
-
-1. **物理删除OSS文件**：删除时调用 FileService 删除OSS文件
-2. **预览图生成**：实现视频封面和图片缩略图异步生成
-3. **文件关联**：充分利用 `file_id` 关联 infra_file 表
-4. **批量操作优化**：优化批量删除性能
-
-### 8.3 整体评价
-
-**设计评分：8.5/10**
-
-- ✅ 架构清晰，分层合理
-- ✅ 路径设计合理，支持多租户
-- ✅ 配额管理完善
-- ⚠️ 物理删除功能缺失
-- ⚠️ 预览图功能未实现
-- ⚠️ 部分字段未充分利用
-
diff --git a/yudao-module-tik/LOGIC_ANALYSIS.md b/yudao-module-tik/LOGIC_ANALYSIS.md
deleted file mode 100644
index 4083fa7fb4..0000000000
--- a/yudao-module-tik/LOGIC_ANALYSIS.md
+++ /dev/null
@@ -1,87 +0,0 @@
-# 文件上传逻辑分析与问题
-
-## 🔴 严重问题：路径不一致
-
-### 问题描述
-
-当前代码存在**路径不一致**的严重问题：
-
-1. **FileService.createFile()** 内部调用 `generateUploadPath()` 生成路径
-   - 使用 `System.currentTimeMillis()` 作为时间戳
-   - 实际存储路径：`{baseDirectory}/{yyyyMMdd}/{filename}_{timestamp1}.ext`
-
-2. **我们手动调用 generateFullFilePath()** 生成路径
-   - 也使用 `System.currentTimeMillis()` 作为时间戳
-   - 但调用时间不同，时间戳可能不同：`{baseDirectory}/{yyyyMMdd}/{filename}_{timestamp2}.ext`
-
-3. **结果**：`filePath` 字段保存的路径 ≠ 实际 OSS 存储路径
-   - 导致删除文件时无法找到正确的文件
-   - 导致路径查询不准确
-
-### 时间戳不一致示例
-
-```
-FileService.createFile() 调用时间：2025-01-15 10:30:45.123
-  → 生成时间戳：1736905845123
-  → 实际路径：video/20250115/file_1736905845123.mp4
-
-generateFullFilePath() 调用时间：2025-01-15 10:30:45.125（2毫秒后）
-  → 生成时间戳：1736905845125
-  → 保存路径：video/20250115/file_1736905845125.mp4
-
-❌ 路径不匹配！
-```
-
-## 📋 冗余代码分析
-
-### 1. generateFullFilePath() 方法
-- **状态**：冗余
-- **原因**：完全复制了 `FileService.generateUploadPath()` 的逻辑
-- **问题**：时间戳不一致导致路径不匹配
-
-### 2. extractPathFromUrl() 方法
-- **状态**：未使用
-- **原因**：创建了但从未调用
-- **建议**：删除或实现使用
-
-## ✅ 解决方案
-
-### 方案1：从 infra_file 表查询 path（推荐）
-
-**优点**：
-- 路径100%准确
-- 可以关联 file_id
-- 逻辑清晰
-
-**实现**：
-```java
-// 上传后，通过 URL 查询 infra_file 表获取 path
-FileDO infraFile = fileMapper.selectOne(
-    new LambdaQueryWrapperX<FileDO>()
-        .eq(FileDO::getUrl, fileUrl)
-        .orderByDesc(FileDO::getCreateTime)
-        .last("LIMIT 1")
-);
-String filePath = infraFile != null ? infraFile.getPath() : null;
-```
-
-### 方案2：从 URL 中提取 path
-
-**优点**：
-- 不需要查询数据库
-- 性能好
-
-**缺点**：
-- URL 可能包含域名、查询参数
-- 提取逻辑复杂，可能不准确
-
-### 方案3：修改 FileApi 返回 path（不推荐）
-
-**缺点**：
-- 需要修改框架代码
-- 影响其他模块
-
-## 🎯 推荐实现
-
-**使用方案1**：从 infra_file 表查询 path，确保路径100%准确。
-
diff --git a/yudao-module-tik/LOGIC_REVIEW.md b/yudao-module-tik/LOGIC_REVIEW.md
deleted file mode 100644
index f9541c975f..0000000000
--- a/yudao-module-tik/LOGIC_REVIEW.md
+++ /dev/null
@@ -1,111 +0,0 @@
-# 文件上传逻辑检查报告
-
-## ✅ 已修复的问题
-
-### 1. 路径不一致问题（已修复）
-
-**问题**：
-- `FileService.createFile()` 和 `generateFullFilePath()` 使用不同的时间戳
-- 导致 `filePath` 和实际 OSS 路径不匹配
-
-**修复方案**：
-- 从 `infra_file` 表查询实际路径（通过 URL + 文件大小）
-- 确保路径100%准确
-- 兜底方案：从 URL 提取路径
-
-**代码位置**：
-```java
-// 从 infra_file 表查询实际的文件路径（确保路径100%准确）
-String filePath = getFilePathFromInfraFile(fileUrl, file.getSize());
-if (StrUtil.isBlank(filePath)) {
-    // 如果查询失败，从URL中提取路径（兜底方案）
-    filePath = extractPathFromUrl(fileUrl);
-}
-```
-
-### 2. 冗余代码清理
-
-**已删除**：
-- `generateFullFilePath()` 方法（已删除，不再需要手动生成路径）
-
-**保留**：
-- `extractPathFromUrl()` 方法（作为兜底方案，在删除文件时也会用到）
-
-## 📊 当前逻辑流程
-
-```
-1. 校验文件分类
-   ↓
-2. 校验配额
-   ↓
-3. 获取OSS基础目录
-   ↓
-4. 读取文件内容
-   ↓
-5. 上传到OSS（FileService.createFile）
-   - FileService 自动生成路径并保存到 infra_file 表
-   - 返回 fileUrl
-   ↓
-6. 从 infra_file 表查询实际路径（✅ 确保准确）
-   - 通过 URL + 文件大小精确匹配
-   - 兜底：从 URL 提取路径
-   ↓
-7. 获取OSS根路径
-   ↓
-8. 保存文件记录到 tik_user_file 表
-   - file_path: 从 infra_file 表查询的准确路径
-   - file_url: FileService 返回的 URL
-   ↓
-9. 更新配额
-```
-
-## ✅ 逻辑可行性检查
-
-### 1. 路径准确性 ✅
-- **方案**：从 `infra_file` 表查询
-- **准确性**：100%（直接使用 FileService 保存的路径）
-- **性能**：一次数据库查询，可接受
-
-### 2. 兜底方案 ✅
-- **方案**：从 URL 提取路径
-- **适用场景**：查询失败时使用
-- **准确性**：中等（URL 可能包含域名和查询参数）
-
-### 3. 文件删除 ✅
-- **当前**：使用 `file_path` 字段
-- **准确性**：高（路径来自 infra_file 表）
-- **TODO**：实现物理删除 OSS 文件
-
-## 🎯 优化建议
-
-### 1. 关联 file_id（可选）
-
-如果后续需要关联 `infra_file` 表，可以在查询时保存 `file_id`：
-
-```java
-FileDO infraFile = fileMapper.selectOne(...);
-if (infraFile != null) {
-    userFile.setFileId(infraFile.getId());  // 关联 infra_file 表
-    filePath = infraFile.getPath();
-}
-```
-
-### 2. 性能优化（可选）
-
-如果担心查询性能，可以：
-- 添加缓存（URL → path 的映射）
-- 或者：直接使用 URL 提取路径（但准确性降低）
-
-## 📝 总结
-
-**当前逻辑**：
-- ✅ 路径准确性：100%（从 infra_file 表查询）
-- ✅ 代码简洁：删除了冗余的路径生成逻辑
-- ✅ 兜底方案：URL 提取路径
-- ✅ 可行性：完全可行
-
-**建议**：
-- 当前实现已经是最优方案
-- 路径准确性有保障
-- 代码逻辑清晰，无冗余
-
diff --git a/yudao-module-tik/UPLOAD_STRATEGY.md b/yudao-module-tik/UPLOAD_STRATEGY.md
deleted file mode 100644
index 5200358283..0000000000
--- a/yudao-module-tik/UPLOAD_STRATEGY.md
+++ /dev/null
@@ -1,76 +0,0 @@
-# 文件上传策略分析
-
-## 🎯 业界成熟方案：先上传OSS，再存数据库
-
-### 方案对比
-
-| 方案 | 优点 | 缺点 | 适用场景 |
-|------|------|------|----------|
-| **先上传OSS，再存数据库** ✅ | 1. OSS上传失败不影响数据库<br>2. 数据库事务可快速回滚<br>3. 用户体验好（文件已上传）<br>4. 孤立文件可定时清理 | 1. 数据库失败会产生孤立文件<br>2. 需要清理机制 | **推荐方案**（业界主流） |
-| 先存数据库，再上传OSS | 1. 数据库失败不会上传OSS<br>2. 不会产生孤立文件 | 1. OSS上传失败需要回滚数据库<br>2. 数据库事务时间长<br>3. 用户体验差 | 不推荐 |
-
-### 为什么选择"先上传OSS，再存数据库"？
-
-1. **性能优势**
-   - OSS上传是外部服务调用，不应该阻塞数据库事务
-   - 数据库事务时间短，减少锁竞争
-
-2. **可靠性优势**
-   - OSS上传失败，直接返回错误，不产生脏数据
-   - 数据库保存失败，OSS文件可以后续清理（定时任务）
-
-3. **用户体验优势**
-   - 文件已上传成功，即使数据库失败，文件还在
-   - 可以重试数据库保存，无需重新上传
-
-4. **业界实践**
-   - 阿里云、腾讯云、AWS 等主流云服务都推荐此方案
-   - 大多数开源项目采用此方案
-
-### 当前实现方案
-
-```
-1. 校验（文件分类、配额）
-   ↓
-2. 读取文件内容
-   ↓
-3. 上传到OSS（FileService.createFile）
-   - 成功：返回 fileUrl 和 filePath
-   - 失败：直接抛出异常，不保存数据库
-   ↓
-4. 保存数据库（事务中）
-   - 成功：返回文件ID
-   - 失败：删除OSS文件，抛出异常
-   ↓
-5. 更新配额
-```
-
-### 异常处理
-
-1. **OSS上传失败**
-   - 直接抛出异常，不保存数据库
-   - 用户可重试上传
-
-2. **数据库保存失败**
-   - 删除已上传的OSS文件（清理）
-   - 抛出异常，用户可重试
-
-3. **孤立文件清理**
-   - 定时任务清理未关联数据库的OSS文件
-   - 基于 infra_file 表的创建时间判断
-
-### 优化建议
-
-1. **异步清理孤立文件**
-   - 定时任务扫描 infra_file 表
-   - 删除超过7天未关联 tik_user_file 的文件
-
-2. **重试机制**
-   - 数据库保存失败时，记录重试队列
-   - 后台任务重试保存
-
-3. **监控告警**
-   - 监控OSS上传失败率
-   - 监控数据库保存失败率
-   - 监控孤立文件数量
-
diff --git a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/enmus/ErrorCodeConstants.java b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/enmus/ErrorCodeConstants.java
index c4c146b880..b33a2e548c 100644
--- a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/enmus/ErrorCodeConstants.java
+++ b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/enmus/ErrorCodeConstants.java
@@ -25,4 +25,12 @@ public interface ErrorCodeConstants {
     ErrorCode FILE_GROUP_NAME_DUPLICATE = new ErrorCode(1_030_000_012, "分组名称重复");
     ErrorCode FILE_GROUP_NOT_BELONG_TO_USER = new ErrorCode(1_030_000_013, "分组不属于当前用户");
 
+    // ========== 配音管理 1-030-001-000 ==========
+    ErrorCode VOICE_NOT_EXISTS = new ErrorCode(1_030_001_001, "配音不存在");
+    ErrorCode VOICE_NAME_DUPLICATE = new ErrorCode(1_030_001_002, "配音名称重复");
+    ErrorCode VOICE_FILE_NOT_EXISTS = new ErrorCode(1_030_001_003, "音频文件不存在");
+    ErrorCode VOICE_TRANSCRIBE_FAILED = new ErrorCode(1_030_001_004, "语音识别失败");
+    ErrorCode VOICE_TTS_FAILED = new ErrorCode(1_030_001_005, "语音合成失败");
+    ErrorCode LATENTSYNC_SUBMIT_FAILED = new ErrorCode(1_030_001_101, "口型同步任务提交失败");
+
 }
diff --git a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/file/service/TikUserFileServiceImpl.java b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/file/service/TikUserFileServiceImpl.java
index 6fddcada83..de1578dec1 100644
--- a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/file/service/TikUserFileServiceImpl.java
+++ b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/file/service/TikUserFileServiceImpl.java
@@ -5,13 +5,22 @@ import cn.hutool.core.io.IoUtil;
 import cn.hutool.core.util.StrUtil;
 import cn.iocoder.yudao.framework.common.pojo.PageResult;
 import cn.iocoder.yudao.framework.common.util.collection.CollectionUtils;
+import cn.iocoder.yudao.framework.common.util.http.HttpUtils;
 import cn.iocoder.yudao.framework.common.util.object.BeanUtils;
 import cn.iocoder.yudao.framework.security.core.util.SecurityFrameworkUtils;
 import cn.iocoder.yudao.framework.tenant.core.context.TenantContextHolder;
 import cn.iocoder.yudao.module.infra.api.file.FileApi;
 import cn.iocoder.yudao.module.infra.dal.dataobject.file.FileDO;
 import cn.iocoder.yudao.module.infra.dal.mysql.file.FileMapper;
+import cn.iocoder.yudao.module.infra.framework.file.core.client.FileClient;
 import cn.iocoder.yudao.module.infra.framework.file.core.utils.FileTypeUtils;
+import cn.iocoder.yudao.module.infra.service.file.FileConfigService;
+import cn.hutool.core.date.LocalDateTimeUtil;
+import cn.hutool.core.io.FileUtil;
+import cn.hutool.core.lang.Assert;
+import cn.hutool.crypto.digest.DigestUtil;
+
+import static cn.hutool.core.date.DatePattern.PURE_DATE_PATTERN;
 import cn.iocoder.yudao.framework.mybatis.core.query.LambdaQueryWrapperX;
 import cn.iocoder.yudao.module.tik.file.dal.dataobject.TikUserFileDO;
 import cn.iocoder.yudao.module.tik.file.dal.mysql.TikUserFileMapper;
@@ -60,6 +69,9 @@ public class TikUserFileServiceImpl implements TikUserFileService {
     @Resource
     private FileMapper fileMapper;
 
+    @Resource
+    private FileConfigService fileConfigService;
+
     @Override
     public Long uploadFile(MultipartFile file, String fileCategory, String coverBase64) {
         Long userId = SecurityFrameworkUtils.getLoginUserId();
@@ -86,28 +98,52 @@ public class TikUserFileServiceImpl implements TikUserFileService {
             throw exception(FILE_NOT_EXISTS, "文件读取失败");
         }
 
-        // ========== 第二阶段：上传到OSS（不在事务中，优先执行） ==========
-        // 5. 上传文件到OSS（FileService会自动处理文件名，添加日期前缀和时间戳后缀）
-        // FileService.createFile 会自动生成路径：{baseDirectory}/{yyyyMMdd}/{filename}_{timestamp}.ext
-        // 注意：FileService 内部会使用原始文件名，并自动添加时间戳后缀保证唯一性
+        // ========== 第二阶段：上传到OSS并保存文件记录（不在事务中，优先执行） ==========
+        // 采用业界成熟方案：直接使用 fileMapper.insert() 获取文件ID，避免通过 URL 查询
         String fileUrl;
         String filePath;
-        Long infraFileId = null; // 用于失败时删除OSS文件
+        Long infraFileId;
+        
         try {
-            fileUrl = fileApi.createFile(fileContent, file.getOriginalFilename(),
-                    baseDirectory, file.getContentType());
-
-            // 6. 从 infra_file 表查询实际的文件路径（确保路径100%准确）
-            // 因为 FileService 已经保存了文件记录到 infra_file 表，我们可以通过 URL 查询获取准确的 path
-            FileDO infraFile = getInfraFileByUrl(fileUrl, file.getSize());
-            if (infraFile != null) {
-                filePath = infraFile.getPath();
-                infraFileId = infraFile.getId(); // 保存 infra_file.id，用于失败时删除
-            } else {
-                // 如果查询失败，从URL中提取路径（兜底方案）
-                filePath = extractPathFromUrl(fileUrl);
-                log.warn("[uploadFile][无法从infra_file表查询路径，使用URL提取，URL({})]", fileUrl);
+            // 1. 处理文件名和类型
+            String fileName = file.getOriginalFilename();
+            String fileType = file.getContentType();
+            if (StrUtil.isEmpty(fileType)) {
+                fileType = FileTypeUtils.getMineType(fileContent, fileName);
             }
+            if (StrUtil.isEmpty(fileName)) {
+                fileName = DigestUtil.sha256Hex(fileContent);
+            }
+            if (StrUtil.isEmpty(FileUtil.extName(fileName))) {
+                String extension = FileTypeUtils.getExtension(fileType);
+                if (StrUtil.isNotEmpty(extension)) {
+                    fileName = fileName + "." + extension;
+                }
+            }
+
+            // 2. 生成上传路径（与 FileService 保持一致）
+            filePath = generateUploadPath(fileName, baseDirectory);
+
+            // 3. 上传到OSS
+            FileClient client = fileConfigService.getMasterFileClient();
+            Assert.notNull(client, "客户端(master) 不能为空");
+            String presignedUrl = client.upload(fileContent, filePath, fileType);
+
+            // 3.1 移除预签名URL中的签名参数，获取基础URL（用于存储）
+            fileUrl = HttpUtils.removeUrlQuery(presignedUrl);
+
+            // 4. 保存到 infra_file 表，直接获取文件ID（MyBatis Plus 会自动填充自增ID）
+            FileDO infraFile = new FileDO()
+                    .setConfigId(client.getId())
+                    .setName(fileName)
+                    .setPath(filePath)
+                    .setUrl(fileUrl)
+                    .setType(fileType)
+                    .setSize((int) file.getSize());
+            fileMapper.insert(infraFile);
+            infraFileId = infraFile.getId(); // MyBatis Plus 会自动填充自增ID
+            
+            log.info("[uploadFile][文件上传成功，文件编号({})，路径({})]", infraFileId, filePath);
         } catch (Exception e) {
             log.error("[uploadFile][上传OSS失败]", e);
             throw exception(FILE_NOT_EXISTS, "上传OSS失败：" + e.getMessage());
@@ -130,7 +166,13 @@ public class TikUserFileServiceImpl implements TikUserFileService {
     @Transactional(rollbackFor = Exception.class)
     public Long saveFileRecord(Long userId, MultipartFile file, String fileCategory,
                                String fileUrl, String filePath, String coverBase64, String baseDirectory, Long infraFileId) {
-        // 7. 处理视频封面（如果有前端传递的 base64 封面，先处理封面再插入主记录）
+        // 7. 验证 infraFileId 不为空（必须在保存记录之前检查）
+        if (infraFileId == null) {
+            log.error("[saveFileRecord][infra_file.id 为空，无法保存文件记录，用户({})，URL({})]", userId, fileUrl);
+            throw exception(FILE_NOT_EXISTS, "文件记录保存失败：无法获取文件ID");
+        }
+
+        // 8. 处理视频封面（如果有前端传递的 base64 封面，先处理封面再插入主记录）
         String coverUrl = null;
         if (StrUtil.isNotBlank(coverBase64) && StrUtil.containsIgnoreCase(file.getContentType(), "video")) {
             try {
@@ -162,7 +204,8 @@ public class TikUserFileServiceImpl implements TikUserFileService {
                         
                         // 严格验证：确保返回的是有效的 URL，而不是 base64 字符串
                         if (StrUtil.isNotBlank(uploadedUrl) && !uploadedUrl.equals(coverBase64) && !uploadedUrl.contains("data:image")) {
-                            coverUrl = uploadedUrl;
+                            // 移除预签名URL中的签名参数，获取基础URL（用于存储）
+                            coverUrl = HttpUtils.removeUrlQuery(uploadedUrl);
                             log.info("[saveFileRecord][视频封面上传成功，封面URL({})]", coverUrl);
                         } else {
                             log.error("[saveFileRecord][视频封面上传返回无效URL，跳过保存封面。返回URL: {}", uploadedUrl);
@@ -177,7 +220,7 @@ public class TikUserFileServiceImpl implements TikUserFileService {
             }
         }
 
-        // 8. 创建文件记录（保存完整路径，包含封面URL和Base64）
+        // 9. 创建文件记录（保存完整路径，包含封面URL和Base64）
         TikUserFileDO userFile = new TikUserFileDO()
                 .setUserId(userId)
                 .setFileId(infraFileId)  // 关联infra_file表，用于后续通过FileService管理文件
@@ -191,11 +234,12 @@ public class TikUserFileServiceImpl implements TikUserFileService {
                 .setCoverBase64(StrUtil.isNotBlank(coverBase64) ? coverBase64 : null);  // 保存原始base64数据（如果有）
         userFileMapper.insert(userFile);
 
-        // 9. 更新配额
+        // 10. 更新配额
         quotaService.increaseUsedStorage(userId, file.getSize());
 
-        log.info("[saveFileRecord][用户({})保存文件记录成功，文件编号({})]", userId, userFile.getId());
-        return userFile.getId();
+        log.info("[saveFileRecord][用户({})保存文件记录成功，文件编号({})，infra文件编号({})]", userId, userFile.getId(), infraFileId);
+        // 返回 infra_file.id，因为创建配音等操作需要使用 infra_file.id
+        return infraFileId;
     }
 
     /**
@@ -221,31 +265,41 @@ public class TikUserFileServiceImpl implements TikUserFileService {
     }
 
     /**
-     * 从 infra_file 表查询文件信息（返回完整对象，包含 id）
+     * 生成上传路径（与 FileService 保持一致）
+     * 格式：{directory}/{yyyyMMdd}/{filename}_{timestamp}.ext
      */
-    private FileDO getInfraFileByUrl(String fileUrl, long fileSize) {
-        if (StrUtil.isBlank(fileUrl)) {
-            return null;
+    private String generateUploadPath(String name, String directory) {
+        // 1. 生成前缀、后缀
+        String prefix = null;
+        boolean PATH_PREFIX_DATE_ENABLE = true;
+        boolean PATH_SUFFIX_TIMESTAMP_ENABLE = true;
+        
+        if (PATH_PREFIX_DATE_ENABLE) {
+            prefix = LocalDateTimeUtil.format(LocalDateTimeUtil.now(), PURE_DATE_PATTERN);
         }
-        try {
-            // 移除URL中的查询参数（如果有）
-            String cleanUrl = fileUrl;
-            if (fileUrl.contains("?")) {
-                cleanUrl = fileUrl.substring(0, fileUrl.indexOf("?"));
+        String suffix = null;
+        if (PATH_SUFFIX_TIMESTAMP_ENABLE) {
+            suffix = String.valueOf(System.currentTimeMillis());
+        }
+
+        // 2.1 先拼接 suffix 后缀
+        if (StrUtil.isNotEmpty(suffix)) {
+            String ext = FileUtil.extName(name);
+            if (StrUtil.isNotEmpty(ext)) {
+                name = FileUtil.mainName(name) + "_" + suffix + "." + ext;
+            } else {
+                name = name + "_" + suffix;
             }
-            
-            // 通过 URL 和文件大小查询（提高准确性）
-            return fileMapper.selectOne(
-                    new LambdaQueryWrapperX<FileDO>()
-                            .eq(FileDO::getUrl, cleanUrl)
-                            .eq(FileDO::getSize, (int) fileSize)  // FileDO.size 是 Integer
-                            .orderByDesc(FileDO::getCreateTime)
-                            .last("LIMIT 1")
-            );
-        } catch (Exception e) {
-            log.warn("[getInfraFileByUrl][查询infra_file表失败，URL({})]", fileUrl, e);
         }
-        return null;
+        // 2.2 再拼接 prefix 前缀
+        if (StrUtil.isNotEmpty(prefix)) {
+            name = prefix + "/" + name;
+        }
+        // 2.3 最后拼接 directory 目录
+        if (StrUtil.isNotEmpty(directory)) {
+            name = directory + "/" + name;
+        }
+        return name;
     }
 
     @Override
@@ -466,16 +520,28 @@ public class TikUserFileServiceImpl implements TikUserFileService {
             return null;
         }
         try {
+            // 移除URL中的查询参数（签名参数等）
+            String cleanUrl = url;
+            if (url.contains("?")) {
+                cleanUrl = url.substring(0, url.indexOf("?"));
+            }
+            
             // 如果URL包含域名，提取路径部分
-            if (url.contains("://")) {
-                int pathStart = url.indexOf("/", url.indexOf("://") + 3);
+            if (cleanUrl.contains("://")) {
+                int pathStart = cleanUrl.indexOf("/", cleanUrl.indexOf("://") + 3);
                 if (pathStart > 0) {
-                    return url.substring(pathStart);
+                    String fullPath = cleanUrl.substring(pathStart);
+                    // 路径可能包含 bucket 名称，需要提取实际的文件路径
+                    // 例如：/bucket-name/user-id/tenant-id/voice/20251117/file.wav
+                    // 实际 path 可能是：user-id/tenant-id/voice/20251117/file.wav
+                    // 但数据库中的 path 格式是：voice/20251117/file_timestamp.wav
+                    // 所以我们需要找到包含日期格式的部分（yyyyMMdd）
+                    return fullPath;
                 }
             }
-            // 如果已经是路径格式，直接返回
-            if (url.startsWith("/")) {
-                return url;
+            // 如果已经是路径格式，直接返回（去除查询参数）
+            if (cleanUrl.startsWith("/")) {
+                return cleanUrl;
             }
         } catch (Exception e) {
             log.warn("[extractPathFromUrl][从URL提取路径失败，URL({})]", url, e);
diff --git a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/tikhup/service/TikFileTransCharacters.java b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/tikhup/service/TikFileTransCharacters.java
index 01b8ad1cc4..4fc4d19eec 100644
--- a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/tikhup/service/TikFileTransCharacters.java
+++ b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/tikhup/service/TikFileTransCharacters.java
@@ -73,7 +73,7 @@ public class TikFileTransCharacters {
         // 设置是否输出词信息，默认为false，开启时需要设置version为4.0及以上。
         taskObject.put(KEY_ENABLE_WORDS, true);
         String task = taskObject.toJSONString();
-        System.out.println(task);
+        System.out.println("[TikFileTransCharacters][submitFileTransRequest] 请求参数: " + task);
         // 设置以上JSON字符串为Body参数。
         postRequest.putBodyParameter(KEY_TASK, task);
         // 设置为POST方式的请求。
@@ -85,15 +85,24 @@ public class TikFileTransCharacters {
         String taskId = null;
         try {
             CommonResponse postResponse = client.getCommonResponse(postRequest);
-            System.err.println("提交录音文件识别请求的响应：" + postResponse.getData());
-            if (postResponse.getHttpStatus() == 200) {
+            System.err.println("[TikFileTransCharacters][submitFileTransRequest] 提交录音文件识别请求的响应：" + postResponse.getData());
+            int httpStatus = postResponse.getHttpStatus();
+            System.out.println("[TikFileTransCharacters][submitFileTransRequest] HTTP状态码: " + httpStatus);
+            if (httpStatus == 200) {
                 JSONObject result = JSONObject.parseObject(postResponse.getData());
                 String statusText = result.getString(KEY_STATUS_TEXT);
+                System.out.println("[TikFileTransCharacters][submitFileTransRequest] 状态文本: " + statusText);
                 if (STATUS_SUCCESS.equals(statusText)) {
                     taskId = result.getString(KEY_TASK_ID);
+                    System.out.println("[TikFileTransCharacters][submitFileTransRequest] 任务ID: " + taskId);
+                } else {
+                    System.err.println("[TikFileTransCharacters][submitFileTransRequest] 状态不是SUCCESS，状态文本: " + statusText);
                 }
+            } else {
+                System.err.println("[TikFileTransCharacters][submitFileTransRequest] HTTP状态码不是200，状态码: " + httpStatus + "，响应: " + postResponse.getData());
             }
         } catch (ClientException e) {
+            System.err.println("[TikFileTransCharacters][submitFileTransRequest] 异常: " + e.getMessage());
             e.printStackTrace();
         }
         return taskId;
@@ -120,17 +129,25 @@ public class TikFileTransCharacters {
          * 以轮询的方式进行识别结果的查询，直到服务端返回的状态描述为“SUCCESS”或错误描述，则结束轮询。
          */
         String result = null;
+        int pollCount = 0;
         while (true) {
+            pollCount++;
             try {
+                System.out.println("[TikFileTransCharacters][getFileTransResult] 第" + pollCount + "次轮询，taskId: " + taskId);
                 CommonResponse getResponse = client.getCommonResponse(getRequest);
-                System.err.println("识别查询结果：" + getResponse.getData());
-                if (getResponse.getHttpStatus() != 200) {
+                int httpStatus = getResponse.getHttpStatus();
+                String responseData = getResponse.getData();
+                System.err.println("[TikFileTransCharacters][getFileTransResult] 识别查询结果，HTTP状态码: " + httpStatus + "，响应: " + responseData);
+                if (httpStatus != 200) {
+                    System.err.println("[TikFileTransCharacters][getFileTransResult] HTTP状态码不是200，停止轮询，taskId: " + taskId);
                     break;
                 }
-                JSONObject rootObj = JSONObject.parseObject(getResponse.getData());
+                JSONObject rootObj = JSONObject.parseObject(responseData);
                 String statusText = rootObj.getString(KEY_STATUS_TEXT);
+                System.out.println("[TikFileTransCharacters][getFileTransResult] 状态文本: " + statusText);
                 if (STATUS_RUNNING.equals(statusText) || STATUS_QUEUEING.equals(statusText)) {
                     // 继续轮询，注意设置轮询时间间隔。
+                    System.out.println("[TikFileTransCharacters][getFileTransResult] 任务进行中，等待10秒后继续轮询，taskId: " + taskId);
                     Thread.sleep(10000);
                 }
                 else {
@@ -139,15 +156,22 @@ public class TikFileTransCharacters {
                         result = rootObj.getString(KEY_RESULT);
                         // 状态信息为成功，但没有识别结果，则可能是由于文件里全是静音、噪音等导致识别为空。
                         if(result == null) {
+                            System.out.println("[TikFileTransCharacters][getFileTransResult] 识别成功但结果为空，taskId: " + taskId);
                             result = "";
+                        } else {
+                            System.out.println("[TikFileTransCharacters][getFileTransResult] 识别成功，结果长度: " + result.length() + "，taskId: " + taskId);
                         }
+                    } else {
+                        System.err.println("[TikFileTransCharacters][getFileTransResult] 状态不是SUCCESS，状态文本: " + statusText + "，taskId: " + taskId);
                     }
                     break;
                 }
             } catch (Exception e) {
+                System.err.println("[TikFileTransCharacters][getFileTransResult] 轮询异常，taskId: " + taskId + "，异常信息: " + e.getMessage());
                 e.printStackTrace();
             }
         }
+        System.out.println("[TikFileTransCharacters][getFileTransResult] 轮询结束，taskId: " + taskId + "，结果: " + (result != null ? "非空，长度" + result.length() : "null"));
         return result;
     }
     public static void main(String args[]) throws Exception {
diff --git a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/tikhup/service/TikHupServiceImpl.java b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/tikhup/service/TikHupServiceImpl.java
index 267192e78f..4c68a312d7 100644
--- a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/tikhup/service/TikHupServiceImpl.java
+++ b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/tikhup/service/TikHupServiceImpl.java
@@ -165,17 +165,20 @@ public class TikHupServiceImpl implements TikHupService {
 
     @Override
     public Object videoToCharacters(String fileLink){
+        log.info("[videoToCharacters][开始识别，文件链接({})]", fileLink);
         TikFileTransCharacters tikFileTransCharacters = new TikFileTransCharacters(accessKeyId, accessKeySecret);
-        // 第一步：提交录音文件识别请求，获取任务ID用于后续的识别结果轮询。
         String taskId = tikFileTransCharacters.submitFileTransRequest(appKey, fileLink);
         if (taskId == null) {
+            log.error("[videoToCharacters][提交识别请求失败，taskId为null，fileLink({})]", fileLink);
             return CommonResult.error(500,"录音文件识别请求失败！");
         }
-        // 第二步：根据任务ID轮询识别结果。
+        log.info("[videoToCharacters][提交识别请求成功，taskId({})]", taskId);
         String transResult = tikFileTransCharacters.getFileTransResult(taskId);
         if (transResult == null) {
+            log.error("[videoToCharacters][识别结果查询失败，taskId({})，transResult为null]", taskId);
             return CommonResult.error(501,"录音文件识别请求失败！");
         }
+        log.info("[videoToCharacters][识别成功，taskId({})，结果长度({})]", taskId, transResult.length());
         return CommonResult.success(transResult);
     }
 
@@ -183,30 +186,28 @@ public class TikHupServiceImpl implements TikHupService {
 
     @Override
     public Object videoToCharacters2(List<String> fileLinkList){
-        // 创建转写请求参数
-        TranscriptionParam param =
-                TranscriptionParam.builder()
-                        // 若没有将API Key配置到环境变量中，需将apiKey替换为自己的API Key
-                        .apiKey(apiKey)
-                        .model("paraformer-v1")
-                        // “language_hints”只支持paraformer-v2模型
-                        .parameter("language_hints", new String[]{"zh", "en"})
-                        .fileUrls(fileLinkList)
-                        .build();
+        log.info("[videoToCharacters2][开始识别，文件数量({})，文件URL({})]", 
+                fileLinkList != null ? fileLinkList.size() : 0, fileLinkList);
+        TranscriptionParam param = TranscriptionParam.builder()
+                .apiKey(apiKey)
+                .model("paraformer-v1")
+                .parameter("language_hints", new String[]{"zh", "en"})
+                .fileUrls(fileLinkList)
+                .build();
         try {
             Transcription transcription = new Transcription();
-            // 提交转写请求
             TranscriptionResult result = transcription.asyncCall(param);
-            log.info("RequestId: {}" ,result.getRequestId());
-            // 阻塞等待任务完成并获取结果
+            log.info("[videoToCharacters2][提交转写请求成功，TaskId({})]", result.getTaskId());
             result = transcription.wait(
                     TranscriptionQueryParam.FromTranscriptionParam(param, result.getTaskId()));
-            return CommonResult.success(new GsonBuilder().setPrettyPrinting().create().toJson(result.getOutput()));
+            String outputJson = new GsonBuilder().setPrettyPrinting().create().toJson(result.getOutput());
+            log.info("[videoToCharacters2][识别成功，TaskId({})，结果长度({})]", 
+                    result.getTaskId(), outputJson != null ? outputJson.length() : 0);
+            return CommonResult.success(outputJson);
         } catch (Exception e) {
-            log.error(e.getMessage());
+            log.error("[videoToCharacters2][识别失败，文件URL({})，异常({})]", fileLinkList, e.getMessage(), e);
             return CommonResult.error(500,"录音文件识别请求失败！");
         }
-
     }
 
 
diff --git a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/client/CosyVoiceClient.java b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/client/CosyVoiceClient.java
new file mode 100644
index 0000000000..19e674a5ef
--- /dev/null
+++ b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/client/CosyVoiceClient.java
@@ -0,0 +1,178 @@
+package cn.iocoder.yudao.module.tik.voice.client;
+
+import cn.hutool.core.collection.CollUtil;
+import cn.hutool.core.util.StrUtil;
+import cn.iocoder.yudao.framework.common.exception.ServiceException;
+import cn.iocoder.yudao.module.tik.voice.client.dto.CosyVoiceTtsRequest;
+import cn.iocoder.yudao.module.tik.voice.client.dto.CosyVoiceTtsResult;
+import cn.iocoder.yudao.module.tik.voice.config.CosyVoiceProperties;
+import com.fasterxml.jackson.databind.JsonNode;
+import com.fasterxml.jackson.databind.ObjectMapper;
+import lombok.RequiredArgsConstructor;
+import lombok.extern.slf4j.Slf4j;
+import okhttp3.MediaType;
+import okhttp3.OkHttpClient;
+import okhttp3.Request;
+import okhttp3.RequestBody;
+import okhttp3.Response;
+import org.springframework.stereotype.Component;
+
+import java.nio.charset.StandardCharsets;
+import java.time.Duration;
+import java.util.Base64;
+import java.util.HashMap;
+import java.util.Map;
+import java.util.concurrent.TimeUnit;
+
+import static cn.iocoder.yudao.framework.common.exception.util.ServiceExceptionUtil.exception;
+import static cn.iocoder.yudao.framework.common.exception.util.ServiceExceptionUtil.exception0;
+import static cn.iocoder.yudao.module.tik.enmus.ErrorCodeConstants.VOICE_TTS_FAILED;
+
+/**
+ * CosyVoice 客户端
+ */
+@Slf4j
+@Component
+@RequiredArgsConstructor
+public class CosyVoiceClient {
+
+    private static final MediaType JSON = MediaType.parse("application/json; charset=utf-8");
+
+    private final CosyVoiceProperties properties;
+    private final ObjectMapper objectMapper;
+
+    private volatile OkHttpClient httpClient;
+
+    /**
+     * 调用 CosyVoice TTS 接口
+     */
+    public CosyVoiceTtsResult synthesize(CosyVoiceTtsRequest request) {
+        if (!properties.isEnabled()) {
+            throw exception0(VOICE_TTS_FAILED.getCode(), "未配置 CosyVoice API Key");
+        }
+        if (request == null || StrUtil.isBlank(request.getText())) {
+            throw exception0(VOICE_TTS_FAILED.getCode(), "TTS 文本不能为空");
+        }
+
+        try {
+            String payload = objectMapper.writeValueAsString(buildPayload(request));
+            Request httpRequest = new Request.Builder()
+                    .url(properties.getTtsUrl())
+                    .addHeader("Authorization", "Bearer " + properties.getApiKey())
+                    .addHeader("Content-Type", "application/json")
+                    .post(RequestBody.create(payload.getBytes(StandardCharsets.UTF_8), JSON))
+                    .build();
+
+            try (Response response = getHttpClient().newCall(httpRequest).execute()) {
+                String body = response.body() != null ? response.body().string() : "";
+                if (!response.isSuccessful()) {
+                    log.error("[CosyVoice][TTS失败][status={}, body={}]", response.code(), body);
+                    throw buildException(body);
+                }
+                return parseTtsResult(body, request);
+            }
+        } catch (ServiceException ex) {
+            throw ex;
+        } catch (Exception ex) {
+            log.error("[CosyVoice][TTS异常]", ex);
+            throw exception(VOICE_TTS_FAILED);
+        }
+    }
+
+    private Map<String, Object> buildPayload(CosyVoiceTtsRequest request) {
+        Map<String, Object> payload = new HashMap<>();
+        String model = StrUtil.blankToDefault(request.getModel(), properties.getDefaultModel());
+        payload.put("model", model);
+
+        Map<String, Object> input = new HashMap<>();
+        input.put("text", request.getText());
+        String voiceId = StrUtil.blankToDefault(request.getVoiceId(), properties.getDefaultVoiceId());
+        if (StrUtil.isNotBlank(voiceId)) {
+            input.put("voice", voiceId);
+        }
+        payload.put("input", input);
+
+        Map<String, Object> parameters = new HashMap<>();
+        int sampleRate = request.getSampleRate() != null ? request.getSampleRate() : properties.getSampleRate();
+        parameters.put("sample_rate", sampleRate);
+        String format = StrUtil.blankToDefault(request.getAudioFormat(), properties.getAudioFormat());
+        parameters.put("format", format);
+        if (request.getSpeechRate() != null) {
+            parameters.put("speech_rate", request.getSpeechRate());
+        }
+        if (request.getVolume() != null) {
+            parameters.put("volume", request.getVolume());
+        }
+        if (request.isPreview()) {
+            parameters.put("preview", true);
+        }
+        payload.put("parameters", parameters);
+        return payload;
+    }
+
+    private CosyVoiceTtsResult parseTtsResult(String body, CosyVoiceTtsRequest request) throws Exception {
+        JsonNode root = objectMapper.readTree(body);
+
+        // 错误响应包含 code 字段
+        if (root.has("code")) {
+            String message = root.has("message") ? root.get("message").asText() : body;
+            log.error("[CosyVoice][TTS失败][code={}, message={}]", root.get("code").asText(), message);
+            throw exception0(VOICE_TTS_FAILED.getCode(), message);
+        }
+
+        JsonNode audioNode = root.path("output").path("audio");
+        if (!audioNode.isArray() || audioNode.isEmpty()) {
+            throw exception0(VOICE_TTS_FAILED.getCode(), "CosyVoice 返回的音频为空");
+        }
+
+        JsonNode firstAudio = audioNode.get(0);
+        String content = firstAudio.path("content").asText();
+        if (StrUtil.isBlank(content)) {
+            throw exception0(VOICE_TTS_FAILED.getCode(), "CosyVoice 返回空音频内容");
+        }
+
+        byte[] audioBytes = Base64.getDecoder().decode(content);
+        CosyVoiceTtsResult result = new CosyVoiceTtsResult();
+        result.setAudio(audioBytes);
+        result.setFormat(firstAudio.path("format").asText(StrUtil.blankToDefault(request.getAudioFormat(), properties.getAudioFormat())));
+        result.setSampleRate(firstAudio.path("sample_rate").asInt(request.getSampleRate() != null ? request.getSampleRate() : properties.getSampleRate()));
+        result.setRequestId(root.path("request_id").asText());
+        result.setVoiceId(firstAudio.path("voice").asText(request.getVoiceId()));
+        return result;
+    }
+
+    private OkHttpClient getHttpClient() {
+        if (httpClient == null) {
+            synchronized (this) {
+                if (httpClient == null) {
+                    java.time.Duration connect = defaultDuration(properties.getConnectTimeout(), 10);
+                    java.time.Duration read = defaultDuration(properties.getReadTimeout(), 60);
+                    httpClient = new OkHttpClient.Builder()
+                            .connectTimeout(connect.toMillis(), TimeUnit.MILLISECONDS)
+                            .readTimeout(read.toMillis(), TimeUnit.MILLISECONDS)
+                            .build();
+                }
+            }
+        }
+        return httpClient;
+    }
+
+    private Duration defaultDuration(Duration duration, long seconds) {
+        return duration == null ? Duration.ofSeconds(seconds) : duration;
+    }
+
+    private ServiceException buildException(String body) {
+        try {
+            JsonNode root = objectMapper.readTree(body);
+            String message = CollUtil.getFirst(
+                    CollUtil.newArrayList(
+                            root.path("message").asText(null),
+                            root.path("output").path("message").asText(null)));
+            return exception0(VOICE_TTS_FAILED.getCode(), StrUtil.blankToDefault(message, "CosyVoice 调用失败"));
+        } catch (Exception ignored) {
+            return exception0(VOICE_TTS_FAILED.getCode(), body);
+        }
+    }
+}
+
+
diff --git a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/client/LatentsyncClient.java b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/client/LatentsyncClient.java
new file mode 100644
index 0000000000..56b15914c7
--- /dev/null
+++ b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/client/LatentsyncClient.java
@@ -0,0 +1,141 @@
+package cn.iocoder.yudao.module.tik.voice.client;
+
+import cn.hutool.core.util.StrUtil;
+import cn.iocoder.yudao.framework.common.exception.ServiceException;
+import cn.iocoder.yudao.module.tik.voice.client.dto.LatentsyncSubmitRequest;
+import cn.iocoder.yudao.module.tik.voice.client.dto.LatentsyncSubmitResponse;
+import cn.iocoder.yudao.module.tik.voice.config.LatentsyncProperties;
+import com.fasterxml.jackson.databind.JsonNode;
+import com.fasterxml.jackson.databind.ObjectMapper;
+import lombok.RequiredArgsConstructor;
+import lombok.extern.slf4j.Slf4j;
+import okhttp3.MediaType;
+import okhttp3.OkHttpClient;
+import okhttp3.Request;
+import okhttp3.RequestBody;
+import okhttp3.Response;
+import org.springframework.stereotype.Component;
+
+import java.nio.charset.StandardCharsets;
+import java.time.Duration;
+import java.util.HashMap;
+import java.util.Map;
+import java.util.concurrent.TimeUnit;
+
+import static cn.iocoder.yudao.framework.common.exception.util.ServiceExceptionUtil.exception;
+import static cn.iocoder.yudao.framework.common.exception.util.ServiceExceptionUtil.exception0;
+import static cn.iocoder.yudao.module.tik.enmus.ErrorCodeConstants.LATENTSYNC_SUBMIT_FAILED;
+
+/**
+ * 302AI Latentsync 客户端
+ */
+@Slf4j
+@Component
+@RequiredArgsConstructor
+public class LatentsyncClient {
+
+    private static final MediaType JSON = MediaType.parse("application/json; charset=utf-8");
+
+    private final LatentsyncProperties properties;
+    private final ObjectMapper objectMapper;
+
+    private volatile OkHttpClient httpClient;
+
+    public LatentsyncSubmitResponse submitTask(LatentsyncSubmitRequest request) {
+        if (!properties.isEnabled()) {
+            throw exception0(LATENTSYNC_SUBMIT_FAILED.getCode(), "未配置 Latentsync API Key");
+        }
+        validateRequest(request);
+
+        Map<String, Object> payload = buildPayload(request);
+        try {
+            String body = objectMapper.writeValueAsString(payload);
+            Request httpRequest = new Request.Builder()
+                    .url(properties.getSubmitUrl())
+                    .addHeader("Authorization", "Bearer " + properties.getApiKey())
+                    .addHeader("Content-Type", "application/json")
+                    .post(RequestBody.create(body.getBytes(StandardCharsets.UTF_8), JSON))
+                    .build();
+
+            try (Response response = getHttpClient().newCall(httpRequest).execute()) {
+                String responseBody = response.body() != null ? response.body().string() : "";
+                if (!response.isSuccessful()) {
+                    log.error("[Latentsync][submit failed][status={}, body={}]", response.code(), responseBody);
+                    throw buildException(responseBody);
+                }
+                LatentsyncSubmitResponse submitResponse =
+                        objectMapper.readValue(responseBody, LatentsyncSubmitResponse.class);
+                if (StrUtil.isBlank(submitResponse.getRequestId())) {
+                    log.error("[Latentsync][submit failed][response={}]", responseBody);
+                    throw exception0(LATENTSYNC_SUBMIT_FAILED.getCode(), "Latentsync 返回 requestId 为空");
+                }
+                return submitResponse;
+            }
+        } catch (ServiceException ex) {
+            throw ex;
+        } catch (Exception ex) {
+            log.error("[Latentsync][submit exception]", ex);
+            throw exception(LATENTSYNC_SUBMIT_FAILED);
+        }
+    }
+
+    private void validateRequest(LatentsyncSubmitRequest request) {
+        if (request == null) {
+            throw exception0(LATENTSYNC_SUBMIT_FAILED.getCode(), "请求体不能为空");
+        }
+        if (StrUtil.isBlank(request.getAudioUrl())) {
+            throw exception0(LATENTSYNC_SUBMIT_FAILED.getCode(), "音频地址不能为空");
+        }
+        if (StrUtil.isBlank(request.getVideoUrl())) {
+            throw exception0(LATENTSYNC_SUBMIT_FAILED.getCode(), "视频地址不能为空");
+        }
+        Integer scale = request.getGuidanceScale();
+        if (scale != null && (scale < 1 || scale > 2)) {
+            throw exception0(LATENTSYNC_SUBMIT_FAILED.getCode(), "guidanceScale 取值范围 1-2");
+        }
+    }
+
+    private Map<String, Object> buildPayload(LatentsyncSubmitRequest request) {
+        Map<String, Object> payload = new HashMap<>();
+        payload.put("audio_url", request.getAudioUrl());
+        payload.put("video_url", request.getVideoUrl());
+        Integer scale = request.getGuidanceScale() != null
+                ? request.getGuidanceScale() : properties.getDefaultGuidanceScale();
+        payload.put("guidance_scale", scale);
+        Integer seed = request.getSeed() != null ? request.getSeed() : properties.getDefaultSeed();
+        payload.put("seed", seed);
+        return payload;
+    }
+
+    private OkHttpClient getHttpClient() {
+        if (httpClient == null) {
+            synchronized (this) {
+                if (httpClient == null) {
+                    Duration connect = defaultDuration(properties.getConnectTimeout(), 10);
+                    Duration read = defaultDuration(properties.getReadTimeout(), 60);
+                    httpClient = new OkHttpClient.Builder()
+                            .connectTimeout(connect.toMillis(), TimeUnit.MILLISECONDS)
+                            .readTimeout(read.toMillis(), TimeUnit.MILLISECONDS)
+                            .build();
+                }
+            }
+        }
+        return httpClient;
+    }
+
+    private Duration defaultDuration(Duration duration, long seconds) {
+        return duration == null ? Duration.ofSeconds(seconds) : duration;
+    }
+
+    private ServiceException buildException(String body) {
+        try {
+            JsonNode root = objectMapper.readTree(body);
+            String message = root.path("message").asText(body);
+            return exception0(LATENTSYNC_SUBMIT_FAILED.getCode(), message);
+        } catch (Exception ignored) {
+            return exception0(LATENTSYNC_SUBMIT_FAILED.getCode(), body);
+        }
+    }
+}
+
+
diff --git a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/client/dto/CosyVoiceTtsRequest.java b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/client/dto/CosyVoiceTtsRequest.java
new file mode 100644
index 0000000000..6fcc1f66b9
--- /dev/null
+++ b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/client/dto/CosyVoiceTtsRequest.java
@@ -0,0 +1,54 @@
+package cn.iocoder.yudao.module.tik.voice.client.dto;
+
+import lombok.Builder;
+import lombok.Data;
+
+/**
+ * CosyVoice TTS 请求
+ */
+@Data
+@Builder
+public class CosyVoiceTtsRequest {
+
+    /**
+     * 待合成文本
+     */
+    private String text;
+
+    /**
+     * 声音 ID（可选，默认使用配置）
+     */
+    private String voiceId;
+
+    /**
+     * 模型（默认 cosyvoice-v2）
+     */
+    private String model;
+
+    /**
+     * 语速
+     */
+    private Float speechRate;
+
+    /**
+     * 音量，可选
+     */
+    private Float volume;
+
+    /**
+     * 采样率
+     */
+    private Integer sampleRate;
+
+    /**
+     * 音频格式
+     */
+    private String audioFormat;
+
+    /**
+     * 是否仅用于试听，方便服务侧做限流
+     */
+    private boolean preview;
+}
+
+
diff --git a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/client/dto/CosyVoiceTtsResult.java b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/client/dto/CosyVoiceTtsResult.java
new file mode 100644
index 0000000000..3a100fff4a
--- /dev/null
+++ b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/client/dto/CosyVoiceTtsResult.java
@@ -0,0 +1,37 @@
+package cn.iocoder.yudao.module.tik.voice.client.dto;
+
+import lombok.Data;
+
+/**
+ * CosyVoice TTS 响应
+ */
+@Data
+public class CosyVoiceTtsResult {
+
+    /**
+     * 请求ID
+     */
+    private String requestId;
+
+    /**
+     * 返回的音频格式
+     */
+    private String format;
+
+    /**
+     * 采样率
+     */
+    private Integer sampleRate;
+
+    /**
+     * 音频二进制内容
+     */
+    private byte[] audio;
+
+    /**
+     * 音频所使用的 voiceId
+     */
+    private String voiceId;
+}
+
+
diff --git a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/client/dto/LatentsyncSubmitRequest.java b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/client/dto/LatentsyncSubmitRequest.java
new file mode 100644
index 0000000000..f8eb5db999
--- /dev/null
+++ b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/client/dto/LatentsyncSubmitRequest.java
@@ -0,0 +1,34 @@
+package cn.iocoder.yudao.module.tik.voice.client.dto;
+
+import lombok.Builder;
+import lombok.Data;
+
+/**
+ * Latentsync 任务提交请求
+ */
+@Data
+@Builder
+public class LatentsyncSubmitRequest {
+
+    /**
+     * 音频地址（必填）
+     */
+    private String audioUrl;
+
+    /**
+     * 视频地址（必填）
+     */
+    private String videoUrl;
+
+    /**
+     * 口型约束力度（1-2）
+     */
+    private Integer guidanceScale;
+
+    /**
+     * 随机种子
+     */
+    private Integer seed;
+}
+
+
diff --git a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/client/dto/LatentsyncSubmitResponse.java b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/client/dto/LatentsyncSubmitResponse.java
new file mode 100644
index 0000000000..c749aef99e
--- /dev/null
+++ b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/client/dto/LatentsyncSubmitResponse.java
@@ -0,0 +1,39 @@
+package cn.iocoder.yudao.module.tik.voice.client.dto;
+
+import lombok.Data;
+
+import java.util.Map;
+
+/**
+ * Latentsync 任务提交响应
+ */
+@Data
+public class LatentsyncSubmitResponse {
+
+    /**
+     * 日志内容（官方暂未返回，预留）
+     */
+    private Object logs;
+
+    /**
+     * 指标信息
+     */
+    private Map<String, Object> metrics;
+
+    /**
+     * 队列位置
+     */
+    private Integer queuePosition;
+
+    /**
+     * 任务 ID
+     */
+    private String requestId;
+
+    /**
+     * 当前状态
+     */
+    private String status;
+}
+
+
diff --git a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/config/CosyVoiceProperties.java b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/config/CosyVoiceProperties.java
new file mode 100644
index 0000000000..60b39c4abb
--- /dev/null
+++ b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/config/CosyVoiceProperties.java
@@ -0,0 +1,74 @@
+package cn.iocoder.yudao.module.tik.voice.config;
+
+import cn.hutool.core.util.StrUtil;
+import lombok.Data;
+import org.springframework.boot.context.properties.ConfigurationProperties;
+import org.springframework.stereotype.Component;
+
+import java.time.Duration;
+
+/**
+ * CosyVoice 配置
+ */
+@Data
+@Component
+@ConfigurationProperties(prefix = "yudao.cosyvoice")
+public class CosyVoiceProperties {
+
+    /**
+     * DashScope API Key
+     */
+    private String apiKey;
+
+    /**
+     * 默认模型
+     */
+    private String defaultModel = "cosyvoice-v2";
+
+    /**
+     * 默认 voiceId（可选）
+     */
+    private String defaultVoiceId;
+
+    /**
+     * 默认采样率
+     */
+    private Integer sampleRate = 24000;
+
+    /**
+     * 默认音频格式
+     */
+    private String audioFormat = "wav";
+
+    /**
+     * 试听默认示例文本
+     */
+    private String previewText = "您好，欢迎体验专属音色。";
+
+    /**
+     * TTS 接口地址
+     */
+    private String ttsUrl = "https://dashscope.aliyuncs.com/api/v1/services/audio/tts/speech-synthesis";
+
+    /**
+     * 连接超时时间
+     */
+    private Duration connectTimeout = Duration.ofSeconds(10);
+
+    /**
+     * 读取超时时间
+     */
+    private Duration readTimeout = Duration.ofSeconds(60);
+
+    /**
+     * 是否启用
+     */
+    private boolean enabled = true;
+
+    public boolean isEnabled() {
+        return enabled && StrUtil.isNotBlank(apiKey);
+    }
+
+}
+
+
diff --git a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/config/LatentsyncProperties.java b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/config/LatentsyncProperties.java
new file mode 100644
index 0000000000..b9e8d3b099
--- /dev/null
+++ b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/config/LatentsyncProperties.java
@@ -0,0 +1,78 @@
+package cn.iocoder.yudao.module.tik.voice.config;
+
+import cn.hutool.core.util.StrUtil;
+import lombok.Data;
+import org.springframework.boot.context.properties.ConfigurationProperties;
+import org.springframework.stereotype.Component;
+
+import java.time.Duration;
+
+/**
+ * Latentsync 接口配置
+ */
+@Data
+@Component
+@ConfigurationProperties(prefix = "tik.latentsync")
+public class LatentsyncProperties {
+
+    /**
+     * 302AI API Key（可通过配置覆盖）
+     */
+    private String apiKey = "ab900d8c94094a90aed3e88cdba785c1";
+
+    /**
+     * 默认海外网关
+     */
+    private String baseUrl = "https://api.302.ai";
+
+    /**
+     * 默认国内中转网关
+     */
+    private String domesticBaseUrl = "https://api.302ai.cn";
+
+    /**
+     * 是否优先使用国内网关
+     */
+    private boolean preferDomestic = false;
+
+    /**
+     * 提交任务路径
+     */
+    private String submitPath = "/302/submit/latentsync";
+
+    /**
+     * guidance_scale 默认值（1-2）
+     */
+    private Integer defaultGuidanceScale = 1;
+
+    /**
+     * 随机种子默认值
+     */
+    private Integer defaultSeed = 8888;
+
+    /**
+     * 连接超时时间
+     */
+    private Duration connectTimeout = Duration.ofSeconds(10);
+
+    /**
+     * 读取超时时间
+     */
+    private Duration readTimeout = Duration.ofSeconds(60);
+
+    /**
+     * 是否打开调用
+     */
+    private boolean enabled = true;
+
+    public String getSubmitUrl() {
+        String base = preferDomestic ? domesticBaseUrl : baseUrl;
+        return StrUtil.blankToDefault(base, baseUrl) + submitPath;
+    }
+
+    public boolean isEnabled() {
+        return enabled && StrUtil.isNotBlank(apiKey);
+    }
+}
+
+
diff --git a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/controller/AppTikLatentsyncController.java b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/controller/AppTikLatentsyncController.java
new file mode 100644
index 0000000000..6e432f6284
--- /dev/null
+++ b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/controller/AppTikLatentsyncController.java
@@ -0,0 +1,38 @@
+package cn.iocoder.yudao.module.tik.voice.controller;
+
+import cn.iocoder.yudao.framework.common.pojo.CommonResult;
+import cn.iocoder.yudao.module.tik.voice.service.LatentsyncService;
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikLatentsyncSubmitReqVO;
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikLatentsyncSubmitRespVO;
+import io.swagger.v3.oas.annotations.Operation;
+import io.swagger.v3.oas.annotations.tags.Tag;
+import jakarta.annotation.Resource;
+import jakarta.validation.Valid;
+import org.springframework.validation.annotation.Validated;
+import org.springframework.web.bind.annotation.PostMapping;
+import org.springframework.web.bind.annotation.RequestBody;
+import org.springframework.web.bind.annotation.RequestMapping;
+import org.springframework.web.bind.annotation.RestController;
+
+import static cn.iocoder.yudao.framework.common.pojo.CommonResult.success;
+
+/**
+ * 用户 App - Latentsync 口型同步
+ */
+@Tag(name = "用户 App - Latentsync 口型同步")
+@RestController
+@RequestMapping("/api/tik/latentsync")
+@Validated
+public class AppTikLatentsyncController {
+
+    @Resource
+    private LatentsyncService latentsyncService;
+
+    @PostMapping("/submit")
+    @Operation(summary = "提交 302AI Latentsync 口型任务")
+    public CommonResult<AppTikLatentsyncSubmitRespVO> submitTask(@Valid @RequestBody AppTikLatentsyncSubmitReqVO reqVO) {
+        return success(latentsyncService.submitTask(reqVO));
+    }
+}
+
+
diff --git a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/controller/AppTikUserVoiceController.java b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/controller/AppTikUserVoiceController.java
new file mode 100644
index 0000000000..45b5532806
--- /dev/null
+++ b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/controller/AppTikUserVoiceController.java
@@ -0,0 +1,95 @@
+package cn.iocoder.yudao.module.tik.voice.controller;
+
+import cn.iocoder.yudao.framework.common.pojo.CommonResult;
+import cn.iocoder.yudao.framework.common.pojo.PageResult;
+import cn.iocoder.yudao.module.tik.voice.service.TikUserVoiceService;
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikUserVoiceCreateReqVO;
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikUserVoicePageReqVO;
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikUserVoiceRespVO;
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikUserVoiceUpdateReqVO;
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikVoicePreviewReqVO;
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikVoicePreviewRespVO;
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikVoiceTtsReqVO;
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikVoiceTtsRespVO;
+import io.swagger.v3.oas.annotations.Operation;
+import io.swagger.v3.oas.annotations.Parameter;
+import io.swagger.v3.oas.annotations.tags.Tag;
+import jakarta.annotation.Resource;
+import jakarta.validation.Valid;
+import lombok.extern.slf4j.Slf4j;
+import org.springframework.validation.annotation.Validated;
+import org.springframework.web.bind.annotation.*;
+
+import static cn.iocoder.yudao.framework.common.pojo.CommonResult.success;
+
+/**
+ * 用户 App - 配音管理 Controller
+ *
+ * @author 芋道源码
+ */
+@Tag(name = "用户 App - 配音管理")
+@RestController
+@RequestMapping("/api/tik/voice")
+@Validated
+@Slf4j
+public class AppTikUserVoiceController {
+
+    @Resource
+    private TikUserVoiceService voiceService;
+
+    @PostMapping("/create")
+    @Operation(summary = "创建配音")
+    public CommonResult<Long> createVoice(@Valid @RequestBody AppTikUserVoiceCreateReqVO createReqVO) {
+        return success(voiceService.createVoice(createReqVO));
+    }
+
+    @PutMapping("/update")
+    @Operation(summary = "更新配音")
+    public CommonResult<Boolean> updateVoice(@Valid @RequestBody AppTikUserVoiceUpdateReqVO updateReqVO) {
+        voiceService.updateVoice(updateReqVO);
+        return success(true);
+    }
+
+    @DeleteMapping("/delete")
+    @Operation(summary = "删除配音")
+    @Parameter(name = "id", description = "配音编号", required = true, example = "1")
+    public CommonResult<Boolean> deleteVoice(@RequestParam("id") Long id) {
+        voiceService.deleteVoice(id);
+        return success(true);
+    }
+
+    @GetMapping("/page")
+    @Operation(summary = "分页查询配音列表")
+    public CommonResult<PageResult<AppTikUserVoiceRespVO>> getVoicePage(@Valid AppTikUserVoicePageReqVO pageReqVO) {
+        return success(voiceService.getVoicePage(pageReqVO));
+    }
+
+    @GetMapping("/get")
+    @Operation(summary = "获取单个配音")
+    @Parameter(name = "id", description = "配音编号", required = true, example = "1")
+    public CommonResult<AppTikUserVoiceRespVO> getVoice(@RequestParam("id") Long id) {
+        return success(voiceService.getVoice(id));
+    }
+
+    @PostMapping("/transcribe")
+    @Operation(summary = "手动触发语音识别")
+    @Parameter(name = "id", description = "配音编号", required = true, example = "1")
+    public CommonResult<Boolean> transcribeVoice(@RequestParam("id") Long id) {
+        voiceService.transcribeVoice(id);
+        return success(true);
+    }
+
+    @PostMapping("/tts")
+    @Operation(summary = "CosyVoice 文本转语音")
+    public CommonResult<AppTikVoiceTtsRespVO> synthesizeVoice(@Valid @RequestBody AppTikVoiceTtsReqVO reqVO) {
+        return success(voiceService.synthesizeVoice(reqVO));
+    }
+
+    @PostMapping("/preview")
+    @Operation(summary = "我的音色试听")
+    public CommonResult<AppTikVoicePreviewRespVO> previewVoice(@Valid @RequestBody AppTikVoicePreviewReqVO reqVO) {
+        return success(voiceService.previewVoice(reqVO));
+    }
+
+}
+
diff --git a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/dal/dataobject/TikUserVoiceDO.java b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/dal/dataobject/TikUserVoiceDO.java
new file mode 100644
index 0000000000..4ab397b217
--- /dev/null
+++ b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/dal/dataobject/TikUserVoiceDO.java
@@ -0,0 +1,59 @@
+package cn.iocoder.yudao.module.tik.voice.dal.dataobject;
+
+import cn.iocoder.yudao.framework.tenant.core.db.TenantBaseDO;
+import com.baomidou.mybatisplus.annotation.KeySequence;
+import com.baomidou.mybatisplus.annotation.TableId;
+import com.baomidou.mybatisplus.annotation.TableName;
+import lombok.*;
+
+/**
+ * 用户配音 DO
+ *
+ * @author 芋道源码
+ */
+@TableName("tik_user_voice")
+@KeySequence("tik_user_voice_seq") // 用于 Oracle、PostgreSQL、Kingbase、DB2、H2 数据库的主键自增。如果是 MySQL 等数据库，可不写。
+@Data
+@EqualsAndHashCode(callSuper = true)
+@ToString(callSuper = true)
+@Builder
+@NoArgsConstructor
+@AllArgsConstructor
+public class TikUserVoiceDO extends TenantBaseDO {
+
+    /**
+     * 配音编号
+     */
+    @TableId
+    private Long id;
+    /**
+     * 用户编号
+     */
+    private Long userId;
+    /**
+     * 配音名称
+     */
+    private String name;
+    /**
+     * 音频文件编号（关联 infra_file.id）
+     */
+    private Long fileId;
+    /**
+     * 语音识别内容，为空表示未识别，有值表示已识别
+     */
+    private String transcription;
+    /**
+     * 语言：zh-CN-简体中文，zh-TW-繁體中文，en-US-English
+     */
+    private String language;
+    /**
+     * 音色类型：female-女声，male-男声
+     */
+    private String gender;
+    /**
+     * 备注信息
+     */
+    private String note;
+
+}
+
diff --git a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/dal/mysql/TikUserVoiceMapper.java b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/dal/mysql/TikUserVoiceMapper.java
new file mode 100644
index 0000000000..ae709b0b56
--- /dev/null
+++ b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/dal/mysql/TikUserVoiceMapper.java
@@ -0,0 +1,26 @@
+package cn.iocoder.yudao.module.tik.voice.dal.mysql;
+
+import cn.iocoder.yudao.framework.common.pojo.PageResult;
+import cn.iocoder.yudao.framework.mybatis.core.mapper.BaseMapperX;
+import cn.iocoder.yudao.framework.mybatis.core.query.LambdaQueryWrapperX;
+import cn.iocoder.yudao.module.tik.voice.dal.dataobject.TikUserVoiceDO;
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikUserVoicePageReqVO;
+import org.apache.ibatis.annotations.Mapper;
+
+/**
+ * 用户配音 Mapper
+ *
+ * @author 芋道源码
+ */
+@Mapper
+public interface TikUserVoiceMapper extends BaseMapperX<TikUserVoiceDO> {
+
+    default PageResult<TikUserVoiceDO> selectPage(AppTikUserVoicePageReqVO reqVO) {
+        return selectPage(reqVO, new LambdaQueryWrapperX<TikUserVoiceDO>()
+                .eqIfPresent(TikUserVoiceDO::getUserId, reqVO.getUserId())
+                .likeIfPresent(TikUserVoiceDO::getName, reqVO.getName())
+                .orderByDesc(TikUserVoiceDO::getId));
+    }
+
+}
+
diff --git a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/service/LatentsyncService.java b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/service/LatentsyncService.java
new file mode 100644
index 0000000000..4c14ff0801
--- /dev/null
+++ b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/service/LatentsyncService.java
@@ -0,0 +1,20 @@
+package cn.iocoder.yudao.module.tik.voice.service;
+
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikLatentsyncSubmitReqVO;
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikLatentsyncSubmitRespVO;
+
+/**
+ * Latentsync 口型同步 Service
+ */
+public interface LatentsyncService {
+
+    /**
+     * 提交 302AI Latentsync 任务
+     *
+     * @param reqVO 请求 VO
+     * @return 任务响应
+     */
+    AppTikLatentsyncSubmitRespVO submitTask(AppTikLatentsyncSubmitReqVO reqVO);
+}
+
+
diff --git a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/service/LatentsyncServiceImpl.java b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/service/LatentsyncServiceImpl.java
new file mode 100644
index 0000000000..152956906a
--- /dev/null
+++ b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/service/LatentsyncServiceImpl.java
@@ -0,0 +1,42 @@
+package cn.iocoder.yudao.module.tik.voice.service;
+
+import cn.hutool.core.util.StrUtil;
+import cn.iocoder.yudao.module.tik.voice.client.LatentsyncClient;
+import cn.iocoder.yudao.module.tik.voice.client.dto.LatentsyncSubmitRequest;
+import cn.iocoder.yudao.module.tik.voice.client.dto.LatentsyncSubmitResponse;
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikLatentsyncSubmitReqVO;
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikLatentsyncSubmitRespVO;
+import jakarta.validation.Valid;
+import lombok.RequiredArgsConstructor;
+import org.springframework.stereotype.Service;
+import org.springframework.validation.annotation.Validated;
+
+/**
+ * Latentsync Service 实现
+ */
+@Service
+@Validated
+@RequiredArgsConstructor
+public class LatentsyncServiceImpl implements LatentsyncService {
+
+    private final LatentsyncClient latentsyncClient;
+
+    @Override
+    public AppTikLatentsyncSubmitRespVO submitTask(@Valid AppTikLatentsyncSubmitReqVO reqVO) {
+        LatentsyncSubmitRequest request = LatentsyncSubmitRequest.builder()
+                .audioUrl(StrUtil.trim(reqVO.getAudioUrl()))
+                .videoUrl(StrUtil.trim(reqVO.getVideoUrl()))
+                .guidanceScale(reqVO.getGuidanceScale())
+                .seed(reqVO.getSeed())
+                .build();
+
+        LatentsyncSubmitResponse response = latentsyncClient.submitTask(request);
+        AppTikLatentsyncSubmitRespVO respVO = new AppTikLatentsyncSubmitRespVO();
+        respVO.setRequestId(response.getRequestId());
+        respVO.setStatus(response.getStatus());
+        respVO.setQueuePosition(response.getQueuePosition());
+        return respVO;
+    }
+}
+
+
diff --git a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/service/TikUserVoiceService.java b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/service/TikUserVoiceService.java
new file mode 100644
index 0000000000..c6ebaf5bbf
--- /dev/null
+++ b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/service/TikUserVoiceService.java
@@ -0,0 +1,75 @@
+package cn.iocoder.yudao.module.tik.voice.service;
+
+import cn.iocoder.yudao.framework.common.pojo.PageResult;
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikUserVoiceCreateReqVO;
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikUserVoicePageReqVO;
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikUserVoiceRespVO;
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikUserVoiceUpdateReqVO;
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikVoicePreviewReqVO;
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikVoicePreviewRespVO;
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikVoiceTtsReqVO;
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikVoiceTtsRespVO;
+
+/**
+ * 用户配音 Service 接口
+ *
+ * @author 芋道源码
+ */
+public interface TikUserVoiceService {
+
+    /**
+     * 创建配音（上传文件 + 可选自动识别）
+     *
+     * @param createReqVO 创建请求 VO
+     * @return 配音编号
+     */
+    Long createVoice(AppTikUserVoiceCreateReqVO createReqVO);
+
+    /**
+     * 更新配音信息
+     *
+     * @param updateReqVO 更新请求 VO
+     */
+    void updateVoice(AppTikUserVoiceUpdateReqVO updateReqVO);
+
+    /**
+     * 删除配音
+     *
+     * @param id 配音编号
+     */
+    void deleteVoice(Long id);
+
+    /**
+     * 分页查询
+     *
+     * @param pageReqVO 分页查询条件
+     * @return 配音列表
+     */
+    PageResult<AppTikUserVoiceRespVO> getVoicePage(AppTikUserVoicePageReqVO pageReqVO);
+
+    /**
+     * 获取单个配音
+     *
+     * @param id 配音编号
+     * @return 配音信息
+     */
+    AppTikUserVoiceRespVO getVoice(Long id);
+
+    /**
+     * 手动触发语音识别
+     *
+     * @param id 配音编号
+     */
+    void transcribeVoice(Long id);
+
+    /**
+     * CosyVoice 文本转语音
+     */
+    AppTikVoiceTtsRespVO synthesizeVoice(AppTikVoiceTtsReqVO reqVO);
+
+    /**
+     * 我的音色试听
+     */
+    AppTikVoicePreviewRespVO previewVoice(AppTikVoicePreviewReqVO reqVO);
+}
+
diff --git a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/service/TikUserVoiceServiceImpl.java b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/service/TikUserVoiceServiceImpl.java
new file mode 100644
index 0000000000..faf6058ce8
--- /dev/null
+++ b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/service/TikUserVoiceServiceImpl.java
@@ -0,0 +1,864 @@
+package cn.iocoder.yudao.module.tik.voice.service;
+
+import cn.hutool.core.collection.CollUtil;
+import cn.hutool.core.util.StrUtil;
+import cn.hutool.http.HttpUtil;
+import cn.hutool.json.JSONArray;
+import cn.hutool.json.JSONObject;
+import cn.hutool.json.JSONUtil;
+import cn.iocoder.yudao.framework.common.pojo.CommonResult;
+import cn.iocoder.yudao.framework.common.pojo.PageResult;
+import cn.iocoder.yudao.framework.common.util.collection.CollectionUtils;
+import cn.iocoder.yudao.framework.common.util.object.BeanUtils;
+import cn.iocoder.yudao.framework.security.core.util.SecurityFrameworkUtils;
+import cn.iocoder.yudao.module.infra.api.file.FileApi;
+import cn.iocoder.yudao.module.infra.dal.dataobject.file.FileDO;
+import cn.iocoder.yudao.module.infra.dal.mysql.file.FileMapper;
+import cn.iocoder.yudao.module.tik.file.dal.dataobject.TikUserFileDO;
+import cn.iocoder.yudao.module.tik.file.dal.mysql.TikUserFileMapper;
+import cn.iocoder.yudao.module.tik.file.service.TikUserFileService;
+import cn.iocoder.yudao.module.tik.tikhup.service.TikHupService;
+import cn.iocoder.yudao.framework.mybatis.core.query.LambdaQueryWrapperX;
+import cn.iocoder.yudao.module.tik.voice.client.CosyVoiceClient;
+import cn.iocoder.yudao.module.tik.voice.client.dto.CosyVoiceTtsRequest;
+import cn.iocoder.yudao.module.tik.voice.client.dto.CosyVoiceTtsResult;
+import cn.iocoder.yudao.module.tik.voice.config.CosyVoiceProperties;
+import cn.iocoder.yudao.module.tik.voice.dal.dataobject.TikUserVoiceDO;
+import cn.iocoder.yudao.module.tik.voice.dal.mysql.TikUserVoiceMapper;
+import cn.iocoder.yudao.module.tik.voice.util.ByteArrayMultipartFile;
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikUserVoiceCreateReqVO;
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikUserVoicePageReqVO;
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikUserVoiceRespVO;
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikUserVoiceUpdateReqVO;
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikVoicePreviewReqVO;
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikVoicePreviewRespVO;
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikVoiceTtsReqVO;
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikVoiceTtsRespVO;
+import lombok.extern.slf4j.Slf4j;
+import org.springframework.data.redis.core.StringRedisTemplate;
+import org.springframework.scheduling.annotation.Async;
+import org.springframework.stereotype.Service;
+import org.springframework.transaction.annotation.Transactional;
+import org.springframework.validation.annotation.Validated;
+
+import jakarta.annotation.Resource;
+import java.util.Arrays;
+import java.util.Base64;
+import java.util.Collections;
+import java.util.HashMap;
+import java.util.List;
+import java.util.Map;
+import java.util.concurrent.TimeUnit;
+import java.util.stream.Collectors;
+
+import static cn.iocoder.yudao.framework.common.exception.util.ServiceExceptionUtil.exception;
+import static cn.iocoder.yudao.module.tik.enmus.ErrorCodeConstants.*;
+
+/**
+ * 用户配音 Service 实现类
+ *
+ * @author 芋道源码
+ */
+@Service
+@Validated
+@Slf4j
+public class TikUserVoiceServiceImpl implements TikUserVoiceService {
+
+    @Resource
+    private TikUserVoiceMapper voiceMapper;
+
+    @Resource
+    private FileMapper fileMapper;
+
+    @Resource
+    private TikUserFileMapper userFileMapper;
+
+    @Resource
+    private TikUserFileService tikUserFileService;
+
+    @Resource
+    private FileApi fileApi;
+
+    @Resource
+    private TikHupService tikHupService;
+
+    @Resource
+    private CosyVoiceClient cosyVoiceClient;
+
+    @Resource
+    private CosyVoiceProperties cosyVoiceProperties;
+
+    @Resource
+    private StringRedisTemplate stringRedisTemplate;
+
+    /** 预签名URL过期时间（1小时，单位：秒） */
+    private static final int PRESIGN_URL_EXPIRATION_SECONDS = 3600;
+    private static final String PREVIEW_CACHE_PREFIX = "tik:voice:preview:";
+    private static final String SYNTH_CACHE_PREFIX = "tik:voice:tts:";
+    private static final long PREVIEW_CACHE_TTL_SECONDS = 3600;
+    private static final long SYNTH_CACHE_TTL_SECONDS = 24 * 3600;
+
+    @Override
+    @Transactional(rollbackFor = Exception.class)
+    public Long createVoice(AppTikUserVoiceCreateReqVO createReqVO) {
+        Long userId = SecurityFrameworkUtils.getLoginUserId();
+
+        // 1. 校验文件是否存在且属于voice分类
+        FileDO fileDO = fileMapper.selectById(createReqVO.getFileId());
+        if (fileDO == null) {
+            throw exception(VOICE_FILE_NOT_EXISTS);
+        }
+        
+        // 验证文件分类是否为voice（通过tik_user_file表查询）
+        TikUserFileDO userFile = userFileMapper.selectOne(new LambdaQueryWrapperX<TikUserFileDO>()
+                .eq(TikUserFileDO::getFileId, createReqVO.getFileId())
+                .eq(TikUserFileDO::getFileCategory, "voice")
+                .eq(TikUserFileDO::getUserId, userId));
+        if (userFile == null) {
+            throw exception(VOICE_FILE_NOT_EXISTS, "文件不存在或不属于voice分类");
+        }
+
+        // 2. 校验名称是否重复
+        TikUserVoiceDO existingVoice = voiceMapper.selectOne(new LambdaQueryWrapperX<TikUserVoiceDO>()
+                .eq(TikUserVoiceDO::getUserId, userId)
+                .eq(TikUserVoiceDO::getName, createReqVO.getName())
+                .eq(TikUserVoiceDO::getDeleted, false));
+        if (existingVoice != null) {
+            throw exception(VOICE_NAME_DUPLICATE);
+        }
+
+        // 3. 创建配音记录
+        TikUserVoiceDO voice = new TikUserVoiceDO()
+                .setUserId(userId)
+                .setName(createReqVO.getName())
+                .setFileId(createReqVO.getFileId())
+                .setLanguage(StrUtil.blankToDefault(createReqVO.getLanguage(), "zh-CN"))
+                .setGender(StrUtil.blankToDefault(createReqVO.getGender(), "female"))
+                .setNote(createReqVO.getNote())
+                .setTranscription(null); // 初始为空，表示未识别
+        voiceMapper.insert(voice);
+
+        // 4. 如果开启自动识别，异步执行识别
+        if (Boolean.TRUE.equals(createReqVO.getAutoTranscribe())) {
+            String fileAccessUrl = fileApi.presignGetUrl(fileDO.getUrl(), PRESIGN_URL_EXPIRATION_SECONDS);
+            log.info("[createVoice][开启自动识别，配音编号({})，文件ID({})，预签名URL({})]", 
+                    voice.getId(), fileDO.getId(), fileAccessUrl);
+            asyncTranscribeVoice(voice.getId(), fileAccessUrl);
+        }
+
+        log.info("[createVoice][用户({})创建配音成功，配音编号({})]", userId, voice.getId());
+        return voice.getId();
+    }
+
+    @Override
+    @Transactional(rollbackFor = Exception.class)
+    public void updateVoice(AppTikUserVoiceUpdateReqVO updateReqVO) {
+        Long userId = SecurityFrameworkUtils.getLoginUserId();
+
+        // 1. 校验配音是否存在且属于当前用户
+        TikUserVoiceDO voice = voiceMapper.selectById(updateReqVO.getId());
+        if (voice == null || !voice.getUserId().equals(userId)) {
+            throw exception(VOICE_NOT_EXISTS);
+        }
+
+        // 2. 如果更新名称，校验名称是否重复
+        if (StrUtil.isNotBlank(updateReqVO.getName()) && !updateReqVO.getName().equals(voice.getName())) {
+            TikUserVoiceDO existingVoice = voiceMapper.selectOne(new LambdaQueryWrapperX<TikUserVoiceDO>()
+                    .eq(TikUserVoiceDO::getUserId, userId)
+                    .eq(TikUserVoiceDO::getName, updateReqVO.getName())
+                    .eq(TikUserVoiceDO::getDeleted, false)
+                    .ne(TikUserVoiceDO::getId, updateReqVO.getId()));
+            if (existingVoice != null) {
+                throw exception(VOICE_NAME_DUPLICATE);
+            }
+        }
+
+        // 3. 更新配音信息
+        TikUserVoiceDO updateObj = new TikUserVoiceDO()
+                .setId(updateReqVO.getId());
+        if (StrUtil.isNotBlank(updateReqVO.getName())) {
+            updateObj.setName(updateReqVO.getName());
+        }
+        if (StrUtil.isNotBlank(updateReqVO.getLanguage())) {
+            updateObj.setLanguage(updateReqVO.getLanguage());
+        }
+        if (StrUtil.isNotBlank(updateReqVO.getGender())) {
+            updateObj.setGender(updateReqVO.getGender());
+        }
+        if (updateReqVO.getNote() != null) {
+            updateObj.setNote(updateReqVO.getNote());
+        }
+        if (updateReqVO.getTranscription() != null) {
+            updateObj.setTranscription(updateReqVO.getTranscription());
+        }
+        voiceMapper.updateById(updateObj);
+
+        log.info("[updateVoice][用户({})更新配音成功，配音编号({})]", userId, updateReqVO.getId());
+    }
+
+    @Override
+    @Transactional(rollbackFor = Exception.class)
+    public void deleteVoice(Long id) {
+        Long userId = SecurityFrameworkUtils.getLoginUserId();
+
+        // 1. 校验配音是否存在且属于当前用户
+        TikUserVoiceDO voice = voiceMapper.selectById(id);
+        if (voice == null || !voice.getUserId().equals(userId)) {
+            throw exception(VOICE_NOT_EXISTS);
+        }
+
+        // 2. 删除音频文件（含OSS）
+        TikUserFileDO userFile = userFileMapper.selectOne(new LambdaQueryWrapperX<TikUserFileDO>()
+                .eq(TikUserFileDO::getFileId, voice.getFileId())
+                .eq(TikUserFileDO::getUserId, userId));
+        if (userFile != null) {
+            tikUserFileService.deleteFiles(Collections.singletonList(userFile.getId()));
+        }
+
+        // 3. 逻辑删除配音记录
+        voiceMapper.deleteById(id);
+
+        log.info("[deleteVoice][用户({})删除配音成功，配音编号({})]", userId, id);
+    }
+
+    @Override
+    public PageResult<AppTikUserVoiceRespVO> getVoicePage(AppTikUserVoicePageReqVO pageReqVO) {
+        // 自动填充当前登录用户ID
+        Long userId = SecurityFrameworkUtils.getLoginUserId();
+        pageReqVO.setUserId(userId);
+
+        // 查询配音列表
+        PageResult<TikUserVoiceDO> pageResult = voiceMapper.selectPage(pageReqVO);
+
+        // 批量查询文件信息，避免 N+1 查询
+        Map<Long, FileDO> fileMap = new HashMap<>();
+        if (CollUtil.isNotEmpty(pageResult.getList())) {
+            List<Long> fileIds = pageResult.getList().stream()
+                    .map(TikUserVoiceDO::getFileId)
+                    .distinct()
+                    .collect(Collectors.toList());
+            
+            if (CollUtil.isNotEmpty(fileIds)) {
+                List<FileDO> files = fileMapper.selectBatchIds(fileIds);
+                Map<Long, FileDO> tempFileMap = files.stream()
+                        .collect(Collectors.toMap(FileDO::getId, file -> file));
+                fileMap.putAll(tempFileMap);
+            }
+        }
+
+        // 转换为VO并关联查询文件信息
+        return CollectionUtils.convertPage(pageResult, voice -> {
+            AppTikUserVoiceRespVO vo = BeanUtils.toBean(voice, AppTikUserVoiceRespVO.class);
+
+            // 通过 file_id 关联查询文件URL，并生成预签名URL
+            FileDO fileDO = fileMap.get(voice.getFileId());
+            if (fileDO != null) {
+                // 生成预签名URL（1小时有效期）
+                String presignedUrl = fileApi.presignGetUrl(fileDO.getUrl(), PRESIGN_URL_EXPIRATION_SECONDS);
+                vo.setFileUrl(presignedUrl);
+            }
+
+            return vo;
+        });
+    }
+
+    @Override
+    public AppTikUserVoiceRespVO getVoice(Long id) {
+        Long userId = SecurityFrameworkUtils.getLoginUserId();
+
+        // 1. 查询配音
+        TikUserVoiceDO voice = voiceMapper.selectById(id);
+        if (voice == null || !voice.getUserId().equals(userId)) {
+            throw exception(VOICE_NOT_EXISTS);
+        }
+
+        // 2. 转换为VO并关联查询文件信息
+        AppTikUserVoiceRespVO vo = BeanUtils.toBean(voice, AppTikUserVoiceRespVO.class);
+
+        // 通过 file_id 关联查询文件URL，并生成预签名URL
+        FileDO fileDO = fileMapper.selectById(voice.getFileId());
+        if (fileDO != null) {
+            // 生成预签名URL（1小时有效期）
+            String presignedUrl = fileApi.presignGetUrl(fileDO.getUrl(), PRESIGN_URL_EXPIRATION_SECONDS);
+            vo.setFileUrl(presignedUrl);
+        }
+
+        return vo;
+    }
+
+    @Override
+    @Transactional(rollbackFor = Exception.class)
+    public void transcribeVoice(Long id) {
+        Long userId = SecurityFrameworkUtils.getLoginUserId();
+
+        // 1. 校验配音是否存在且属于当前用户
+        TikUserVoiceDO voice = voiceMapper.selectById(id);
+        if (voice == null || !voice.getUserId().equals(userId)) {
+            throw exception(VOICE_NOT_EXISTS);
+        }
+
+        // 2. 获取文件URL
+        FileDO fileDO = fileMapper.selectById(voice.getFileId());
+        if (fileDO == null) {
+            throw exception(VOICE_FILE_NOT_EXISTS);
+        }
+
+        // 3. 异步执行识别
+        String fileAccessUrl = fileApi.presignGetUrl(fileDO.getUrl(), PRESIGN_URL_EXPIRATION_SECONDS);
+        asyncTranscribeVoice(id, fileAccessUrl);
+    }
+
+    @Override
+    public AppTikVoiceTtsRespVO synthesizeVoice(AppTikVoiceTtsReqVO reqVO) {
+        String finalText = determineSynthesisText(
+                reqVO.getTranscriptionText(),
+                reqVO.getInputText(),
+                false);
+        finalText = appendEmotion(finalText, reqVO.getEmotion());
+
+        String cacheKey = buildCacheKey(SYNTH_CACHE_PREFIX,
+                reqVO.getVoiceId(),
+                reqVO.getFileUrl(),
+                finalText,
+                reqVO.getSpeechRate(),
+                reqVO.getVolume(),
+                reqVO.getEmotion(),
+                reqVO.getAudioFormat(),
+                reqVO.getSampleRate());
+
+        SynthCacheEntry synthCache = getSynthCache(cacheKey);
+        if (synthCache != null) {
+            return buildSynthResponseFromCache(reqVO, synthCache);
+        }
+
+        CosyVoiceTtsResult ttsResult = cosyVoiceClient.synthesize(buildTtsRequest(
+                finalText,
+                reqVO.getVoiceId(),
+                reqVO.getModel(),
+                reqVO.getSpeechRate(),
+                reqVO.getVolume(),
+                reqVO.getSampleRate(),
+                reqVO.getAudioFormat(),
+                false
+        ));
+
+        String format = defaultFormat(ttsResult.getFormat(), reqVO.getAudioFormat());
+        String voiceId = StrUtil.blankToDefault(reqVO.getVoiceId(), cosyVoiceProperties.getDefaultVoiceId());
+        ByteArrayMultipartFile multipartFile = new ByteArrayMultipartFile(
+                "file",
+                buildFileName(voiceId, format),
+                resolveContentType(format),
+                ttsResult.getAudio()
+        );
+        Long fileId = tikUserFileService.uploadFile(multipartFile, "audio", null);
+
+        AppTikVoiceTtsRespVO respVO = new AppTikVoiceTtsRespVO();
+        respVO.setFileId(fileId);
+        respVO.setAudioUrl(tikUserFileService.getAudioPlayUrl(fileId));
+        respVO.setFormat(format);
+        respVO.setSampleRate(ttsResult.getSampleRate());
+        respVO.setRequestId(ttsResult.getRequestId());
+        respVO.setVoiceId(voiceId);
+
+        saveSynthCache(cacheKey, new SynthCacheEntry(
+                Base64.getEncoder().encodeToString(ttsResult.getAudio()),
+                format,
+                ttsResult.getSampleRate(),
+                ttsResult.getRequestId(),
+                voiceId
+        ));
+        return respVO;
+    }
+
+    @Override
+    public AppTikVoicePreviewRespVO previewVoice(AppTikVoicePreviewReqVO reqVO) {
+        String finalText = determineSynthesisText(
+                reqVO.getTranscriptionText(),
+                reqVO.getInputText(),
+                true);
+        finalText = appendEmotion(finalText, reqVO.getEmotion());
+
+        String cacheKey = buildCacheKey(PREVIEW_CACHE_PREFIX,
+                reqVO.getVoiceId(),
+                reqVO.getFileUrl(),
+                finalText,
+                reqVO.getSpeechRate(),
+                reqVO.getVolume(),
+                reqVO.getEmotion(),
+                reqVO.getAudioFormat(),
+                null);
+        PreviewCacheEntry previewCache = getPreviewCache(cacheKey);
+        String voiceId = StrUtil.blankToDefault(reqVO.getVoiceId(), cosyVoiceProperties.getDefaultVoiceId());
+
+        if (previewCache != null) {
+            String cachedUrl = fileApi.presignGetUrl(previewCache.getFileUrl(), PRESIGN_URL_EXPIRATION_SECONDS);
+            return buildPreviewResp(previewCache, cachedUrl, voiceId);
+        }
+
+        CosyVoiceTtsResult ttsResult = cosyVoiceClient.synthesize(buildTtsRequest(
+                finalText,
+                reqVO.getVoiceId(),
+                reqVO.getModel(),
+                reqVO.getSpeechRate(),
+                reqVO.getVolume(),
+                null,
+                reqVO.getAudioFormat(),
+                true
+        ));
+
+        String format = defaultFormat(ttsResult.getFormat(), reqVO.getAudioFormat());
+        voiceId = StrUtil.blankToDefault(reqVO.getVoiceId(), cosyVoiceProperties.getDefaultVoiceId());
+        String objectName = buildFileName(voiceId, format);
+        String fileUrl = fileApi.createFile(ttsResult.getAudio(), objectName, "voice/preview", resolveContentType(format));
+        String presignUrl = fileApi.presignGetUrl(fileUrl, PRESIGN_URL_EXPIRATION_SECONDS);
+
+        PreviewCacheEntry entry = new PreviewCacheEntry(fileUrl, format, ttsResult.getSampleRate(), ttsResult.getRequestId());
+        savePreviewCache(cacheKey, entry);
+        return buildPreviewResp(entry, presignUrl, voiceId);
+    }
+
+    private CosyVoiceTtsRequest buildTtsRequest(String text,
+                                                String voiceId,
+                                                String model,
+                                                Float speechRate,
+                                                Float volume,
+                                                Integer sampleRate,
+                                                String audioFormat,
+                                                boolean preview) {
+        return CosyVoiceTtsRequest.builder()
+                .text(text)
+                .voiceId(voiceId)
+                .model(model)
+                .speechRate(speechRate)
+                .volume(volume)
+                .sampleRate(sampleRate)
+                .audioFormat(audioFormat)
+                .preview(preview)
+                .build();
+    }
+
+    private String defaultFormat(String responseFormat, String requestFormat) {
+        return StrUtil.blankToDefault(responseFormat,
+                StrUtil.blankToDefault(requestFormat, cosyVoiceProperties.getAudioFormat()));
+    }
+
+    private String buildFileName(String voiceId, String format) {
+        String safeVoice = StrUtil.blankToDefault(voiceId, "voice")
+                .replaceAll("[^a-zA-Z0-9_-]", "");
+        return safeVoice + "-" + System.currentTimeMillis() + "." + format;
+    }
+
+    private String resolveContentType(String format) {
+        if ("wav".equalsIgnoreCase(format)) {
+            return "audio/wav";
+        }
+        if ("mp3".equalsIgnoreCase(format)) {
+            return "audio/mpeg";
+        }
+        if ("flac".equalsIgnoreCase(format)) {
+            return "audio/flac";
+        }
+        return "audio/mpeg";
+    }
+
+    private String determineSynthesisText(String transcriptionText, String inputText, boolean allowFallback) {
+        StringBuilder builder = new StringBuilder();
+        if (StrUtil.isNotBlank(transcriptionText)) {
+            builder.append(transcriptionText.trim());
+        }
+        if (StrUtil.isNotBlank(inputText)) {
+            if (builder.length() > 0) {
+                builder.append("\n");
+            }
+            builder.append(inputText.trim());
+        }
+        if (builder.length() > 0) {
+            return builder.toString();
+        }
+        if (allowFallback) {
+            return cosyVoiceProperties.getPreviewText();
+        }
+        throw exception(VOICE_TTS_FAILED, "请提供需要合成的文本内容");
+    }
+
+    private String appendEmotion(String text, String emotion) {
+        if (StrUtil.isBlank(text)) {
+            return text;
+        }
+        if (StrUtil.isBlank(emotion) || "neutral".equalsIgnoreCase(emotion)) {
+            return text;
+        }
+        String emotionLabel = switch (emotion.toLowerCase()) {
+            case "happy" -> "高兴";
+            case "angry" -> "愤怒";
+            case "sad" -> "悲伤";
+            case "scared" -> "害怕";
+            case "disgusted" -> "厌恶";
+            case "surprised" -> "惊讶";
+            default -> emotion;
+        };
+        return "【情感：" + emotionLabel + "】" + text;
+    }
+
+    private String buildCacheKey(String prefix,
+                                 String voiceId,
+                                 String fileUrl,
+                                 String text,
+                                 Float speechRate,
+                                 Float volume,
+                                 String emotion,
+                                 String audioFormat,
+                                 Integer sampleRate) {
+        String identifier = StrUtil.isNotBlank(voiceId)
+                ? voiceId
+                : StrUtil.blankToDefault(fileUrl, "no-voice");
+        String payload = StrUtil.join("|",
+                identifier,
+                text,
+                speechRate != null ? speechRate : "1.0",
+                volume != null ? volume : "0",
+                StrUtil.blankToDefault(emotion, "neutral"),
+                StrUtil.blankToDefault(audioFormat, cosyVoiceProperties.getAudioFormat()),
+                sampleRate != null ? sampleRate : cosyVoiceProperties.getSampleRate());
+        String hash = cn.hutool.crypto.SecureUtil.sha256(payload);
+        return prefix + hash;
+    }
+
+    private PreviewCacheEntry getPreviewCache(String key) {
+        try {
+            String json = stringRedisTemplate.opsForValue().get(key);
+            if (StrUtil.isBlank(json)) {
+                return null;
+            }
+            return JSONUtil.toBean(json, PreviewCacheEntry.class);
+        } catch (Exception ex) {
+            log.warn("[previewVoice][cache read failed][key={}]", key, ex);
+            return null;
+        }
+    }
+
+    private void savePreviewCache(String key, PreviewCacheEntry entry) {
+        try {
+            stringRedisTemplate.opsForValue().set(
+                    key,
+                    JSONUtil.toJsonStr(entry),
+                    PREVIEW_CACHE_TTL_SECONDS,
+                    TimeUnit.SECONDS);
+        } catch (Exception ex) {
+            log.warn("[previewVoice][cache write failed][key={}]", key, ex);
+        }
+    }
+
+    private SynthCacheEntry getSynthCache(String key) {
+        try {
+            String json = stringRedisTemplate.opsForValue().get(key);
+            if (StrUtil.isBlank(json)) {
+                return null;
+            }
+            return JSONUtil.toBean(json, SynthCacheEntry.class);
+        } catch (Exception ex) {
+            log.warn("[synthesizeVoice][cache read failed][key={}]", key, ex);
+            return null;
+        }
+    }
+
+    private void saveSynthCache(String key, SynthCacheEntry entry) {
+        try {
+            stringRedisTemplate.opsForValue().set(
+                    key,
+                    JSONUtil.toJsonStr(entry),
+                    SYNTH_CACHE_TTL_SECONDS,
+                    TimeUnit.SECONDS);
+        } catch (Exception ex) {
+            log.warn("[synthesizeVoice][cache write failed][key={}]", key, ex);
+        }
+    }
+
+    private AppTikVoiceTtsRespVO buildSynthResponseFromCache(AppTikVoiceTtsReqVO reqVO, SynthCacheEntry cache) {
+        byte[] audioBytes = Base64.getDecoder().decode(cache.getAudioBase64());
+        String format = defaultFormat(cache.getFormat(), reqVO.getAudioFormat());
+        String voiceId = StrUtil.blankToDefault(reqVO.getVoiceId(), cache.getVoiceId());
+        ByteArrayMultipartFile multipartFile = new ByteArrayMultipartFile(
+                "file",
+                buildFileName(voiceId, format),
+                resolveContentType(format),
+                audioBytes
+        );
+        Long fileId = tikUserFileService.uploadFile(multipartFile, "audio", null);
+
+        AppTikVoiceTtsRespVO respVO = new AppTikVoiceTtsRespVO();
+        respVO.setFileId(fileId);
+        respVO.setAudioUrl(tikUserFileService.getAudioPlayUrl(fileId));
+        respVO.setFormat(format);
+        respVO.setSampleRate(cache.getSampleRate());
+        respVO.setRequestId(cache.getRequestId());
+        respVO.setVoiceId(voiceId);
+        return respVO;
+    }
+
+    private AppTikVoicePreviewRespVO buildPreviewResp(PreviewCacheEntry entry, String presignUrl, String voiceId) {
+        AppTikVoicePreviewRespVO respVO = new AppTikVoicePreviewRespVO();
+        respVO.setAudioUrl(presignUrl);
+        respVO.setFormat(entry.getFormat());
+        respVO.setSampleRate(entry.getSampleRate());
+        respVO.setRequestId(entry.getRequestId());
+        respVO.setVoiceId(voiceId);
+        return respVO;
+    }
+
+    private static class PreviewCacheEntry {
+        private String fileUrl;
+        private String format;
+        private Integer sampleRate;
+        private String requestId;
+
+        public PreviewCacheEntry() {}
+
+        public PreviewCacheEntry(String fileUrl, String format, Integer sampleRate, String requestId) {
+            this.fileUrl = fileUrl;
+            this.format = format;
+            this.sampleRate = sampleRate;
+            this.requestId = requestId;
+        }
+
+        public String getFileUrl() {
+            return fileUrl;
+        }
+
+        public String getFormat() {
+            return format;
+        }
+
+        public Integer getSampleRate() {
+            return sampleRate;
+        }
+
+        public String getRequestId() {
+            return requestId;
+        }
+    }
+
+    private static class SynthCacheEntry {
+        private String audioBase64;
+        private String format;
+        private Integer sampleRate;
+        private String requestId;
+        private String voiceId;
+
+        public SynthCacheEntry() {}
+
+        public SynthCacheEntry(String audioBase64, String format, Integer sampleRate, String requestId, String voiceId) {
+            this.audioBase64 = audioBase64;
+            this.format = format;
+            this.sampleRate = sampleRate;
+            this.requestId = requestId;
+            this.voiceId = voiceId;
+        }
+
+        public String getAudioBase64() {
+            return audioBase64;
+        }
+
+        public String getFormat() {
+            return format;
+        }
+
+        public Integer getSampleRate() {
+            return sampleRate;
+        }
+
+        public String getRequestId() {
+            return requestId;
+        }
+
+        public String getVoiceId() {
+            return voiceId;
+        }
+    }
+
+    /**
+     * 异步执行语音识别
+     *
+     * @param voiceId 配音编号
+     * @param fileUrl 文件URL
+     */
+    @Async
+    public void asyncTranscribeVoice(Long voiceId, String fileUrl) {
+        try {
+            log.info("[asyncTranscribeVoice][开始识别，配音编号({})，文件URL({})]", voiceId, fileUrl);
+            Object result = tikHupService.videoToCharacters2(Collections.singletonList(fileUrl));
+            
+            // 解析识别结果
+            String transcription = extractTranscription(result);
+            
+            if (StrUtil.isNotBlank(transcription)) {
+                // 更新识别结果
+                TikUserVoiceDO updateObj = new TikUserVoiceDO()
+                        .setId(voiceId)
+                        .setTranscription(transcription);
+                voiceMapper.updateById(updateObj);
+                log.info("[asyncTranscribeVoice][识别成功，配音编号({})，文本长度({})]", voiceId, transcription.length());
+            } else {
+                log.warn("[asyncTranscribeVoice][识别结果为空，配音编号({})，返回码({})]", 
+                        voiceId, result instanceof CommonResult ? ((CommonResult<?>) result).getCode() : "未知");
+            }
+        } catch (Exception e) {
+            log.error("[asyncTranscribeVoice][识别失败，配音编号({})，文件URL({})]", voiceId, fileUrl, e);
+        }
+    }
+
+    /**
+     * 从识别结果中提取文字内容
+     * 根据 TikHupService.videoToCharacters* 的实际返回格式进行解析
+     *
+     * @param result 识别结果
+     * @return 文字内容
+     */
+    private String extractTranscription(Object result) {
+        if (result == null) {
+            return null;
+        }
+
+        try {
+            if (result instanceof CommonResult<?> commonResult) {
+                if (!commonResult.isSuccess()) {
+                    log.warn("[extractTranscription][识别失败，code({})，msg({})]",
+                            commonResult.getCode(), commonResult.getMsg());
+                    return null;
+                }
+                Object data = commonResult.getData();
+                if (data == null) {
+                    return null;
+                }
+                String parsed = parseTranscriptionText(data);
+                if (StrUtil.isNotBlank(parsed)) {
+                    return parsed;
+                }
+                return data.toString();
+            }
+
+            String parsed = parseTranscriptionText(result);
+            if (StrUtil.isNotBlank(parsed)) {
+                return parsed;
+            }
+            return result.toString();
+        } catch (Exception e) {
+            log.warn("[extractTranscription][解析识别结果失败]", e);
+            return null;
+        }
+    }
+
+    private static final List<String> TRANSCRIPTION_TEXT_KEYS =
+            Arrays.asList("text", "sentence", "result", "content", "transcript", "output_text", "display_text");
+
+    private String parseTranscriptionText(Object rawData) {
+        if (rawData == null) {
+            return null;
+        }
+        String rawString = rawData instanceof String ? (String) rawData : JSONUtil.toJsonStr(rawData);
+        if (StrUtil.isBlank(rawString)) {
+            return null;
+        }
+        if (!JSONUtil.isTypeJSON(rawString)) {
+            return rawString;
+        }
+        try {
+            Object json = JSONUtil.parse(rawString);
+            String localText = extractTextFromJson(json);
+            if (StrUtil.isNotBlank(localText)) {
+                return localText;
+            }
+            if (json instanceof JSONObject jsonObject) {
+                JSONArray results = jsonObject.getJSONArray("results");
+                if (CollUtil.isEmpty(results)) {
+                    return null;
+                }
+                Object lastObj = results.get(results.size() - 1);
+                if (!(lastObj instanceof JSONObject lastResult)) {
+                    return null;
+                }
+                String transcriptionUrl = lastResult.getStr("transcription_url");
+                if (StrUtil.isBlank(transcriptionUrl)) {
+                    return null;
+                }
+                StringBuilder builder = new StringBuilder();
+                appendRemoteTranscription(builder, transcriptionUrl);
+                return builder.length() > 0 ? builder.toString().trim() : null;
+            }
+        } catch (Exception e) {
+            log.warn("[parseTranscriptionText][解析Paraformer结果失败]", e);
+        }
+        return rawString;
+    }
+
+    private void appendRemoteTranscription(StringBuilder builder, String transcriptionUrl) {
+        if (StrUtil.isBlank(transcriptionUrl)) {
+            return;
+        }
+        String remoteContent = fetchRemoteTranscription(transcriptionUrl);
+        if (StrUtil.isBlank(remoteContent)) {
+            return;
+        }
+        String remoteText = extractTextFromJson(JSONUtil.parse(remoteContent));
+        if (StrUtil.isNotBlank(remoteText)) {
+            appendLine(builder, remoteText);
+        }
+    }
+
+    private String extractTextFromJson(Object json) {
+        if (json == null) {
+            return null;
+        }
+        StringBuilder builder = new StringBuilder();
+        collectTranscriptionText(json, builder);
+        return builder.length() > 0 ? builder.toString().trim() : null;
+    }
+
+    private String fetchRemoteTranscription(String url) {
+        try {
+            String body = HttpUtil.get(url);
+            if (StrUtil.isNotBlank(body)) {
+                return body;
+            }
+        } catch (Exception e) {
+            log.warn("[fetchRemoteTranscription][下载转写文本失败，url({})]", url, e);
+        }
+        return null;
+    }
+
+    private void collectTranscriptionText(Object node, StringBuilder builder) {
+        if (node == null) {
+            return;
+        }
+        if (node instanceof JSONObject jsonObject) {
+            for (String key : jsonObject.keySet()) {
+                Object value = jsonObject.get(key);
+                if (value == null) {
+                    continue;
+                }
+                if (value instanceof CharSequence && TRANSCRIPTION_TEXT_KEYS.contains(key)) {
+                    appendLine(builder, value.toString());
+                } else if (value instanceof JSONObject || value instanceof JSONArray) {
+                    collectTranscriptionText(value, builder);
+                }
+            }
+        } else if (node instanceof JSONArray jsonArray) {
+            for (Object item : jsonArray) {
+                collectTranscriptionText(item, builder);
+            }
+        }
+    }
+
+    private void appendLine(StringBuilder builder, String line) {
+        String normalized = StrUtil.trim(line);
+        if (StrUtil.isBlank(normalized)) {
+            return;
+        }
+        if (builder.length() > 0) {
+            builder.append('\n');
+        }
+        builder.append(normalized);
+    }
+
+}
+
diff --git a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/util/ByteArrayMultipartFile.java b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/util/ByteArrayMultipartFile.java
new file mode 100644
index 0000000000..f74628dd1f
--- /dev/null
+++ b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/util/ByteArrayMultipartFile.java
@@ -0,0 +1,69 @@
+package cn.iocoder.yudao.module.tik.voice.util;
+
+import org.springframework.util.FileCopyUtils;
+import org.springframework.web.multipart.MultipartFile;
+
+import java.io.ByteArrayInputStream;
+import java.io.File;
+import java.io.IOException;
+import java.io.InputStream;
+
+/**
+ * 仅用于在服务内部上传的内存文件
+ */
+public class ByteArrayMultipartFile implements MultipartFile {
+
+    private final String name;
+    private final String originalFilename;
+    private final String contentType;
+    private final byte[] content;
+
+    public ByteArrayMultipartFile(String name, String originalFilename, String contentType, byte[] content) {
+        this.name = name;
+        this.originalFilename = originalFilename;
+        this.contentType = contentType;
+        this.content = content != null ? content : new byte[0];
+    }
+
+    @Override
+    public String getName() {
+        return name;
+    }
+
+    @Override
+    public String getOriginalFilename() {
+        return originalFilename;
+    }
+
+    @Override
+    public String getContentType() {
+        return contentType;
+    }
+
+    @Override
+    public boolean isEmpty() {
+        return content.length == 0;
+    }
+
+    @Override
+    public long getSize() {
+        return content.length;
+    }
+
+    @Override
+    public byte[] getBytes() {
+        return content;
+    }
+
+    @Override
+    public InputStream getInputStream() {
+        return new ByteArrayInputStream(content);
+    }
+
+    @Override
+    public void transferTo(File dest) throws IOException {
+        FileCopyUtils.copy(content, dest);
+    }
+}
+
+
diff --git a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikLatentsyncSubmitReqVO.java b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikLatentsyncSubmitReqVO.java
new file mode 100644
index 0000000000..f89c55ef56
--- /dev/null
+++ b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikLatentsyncSubmitReqVO.java
@@ -0,0 +1,37 @@
+package cn.iocoder.yudao.module.tik.voice.vo;
+
+import io.swagger.v3.oas.annotations.media.Schema;
+import jakarta.validation.constraints.Max;
+import jakarta.validation.constraints.Min;
+import jakarta.validation.constraints.NotBlank;
+import jakarta.validation.constraints.Size;
+import lombok.Data;
+
+/**
+ * Latentsync 提交请求 VO
+ */
+@Data
+public class AppTikLatentsyncSubmitReqVO {
+
+    @Schema(description = "音频 URL（需公网可访问）", requiredMode = Schema.RequiredMode.REQUIRED,
+            example = "https://example.com/audio.wav")
+    @NotBlank(message = "音频地址不能为空")
+    @Size(max = 1024, message = "音频地址长度不能超过 1024 字符")
+    private String audioUrl;
+
+    @Schema(description = "视频 URL（需公网可访问）", requiredMode = Schema.RequiredMode.REQUIRED,
+            example = "https://example.com/video.mp4")
+    @NotBlank(message = "视频地址不能为空")
+    @Size(max = 1024, message = "视频地址长度不能超过 1024 字符")
+    private String videoUrl;
+
+    @Schema(description = "guidance_scale，范围 1-2（默认 1）", example = "1")
+    @Min(value = 1, message = "guidanceScale 不能小于 1")
+    @Max(value = 2, message = "guidanceScale 不能大于 2")
+    private Integer guidanceScale;
+
+    @Schema(description = "随机种子（默认 8888）", example = "8888")
+    private Integer seed;
+}
+
+
diff --git a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikLatentsyncSubmitRespVO.java b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikLatentsyncSubmitRespVO.java
new file mode 100644
index 0000000000..1cc773afd3
--- /dev/null
+++ b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikLatentsyncSubmitRespVO.java
@@ -0,0 +1,22 @@
+package cn.iocoder.yudao.module.tik.voice.vo;
+
+import io.swagger.v3.oas.annotations.media.Schema;
+import lombok.Data;
+
+/**
+ * Latentsync 提交响应 VO
+ */
+@Data
+public class AppTikLatentsyncSubmitRespVO {
+
+    @Schema(description = "Latentsync 任务 ID", example = "8eed0b9b-6103-4357-a57b-9f135a8c3276")
+    private String requestId;
+
+    @Schema(description = "官方状态，如 IN_QUEUE、PROCESSING、SUCCEEDED", example = "IN_QUEUE")
+    private String status;
+
+    @Schema(description = "当前排队位置", example = "0")
+    private Integer queuePosition;
+}
+
+
diff --git a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikUserVoiceCreateReqVO.java b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikUserVoiceCreateReqVO.java
new file mode 100644
index 0000000000..1aaae3eaa6
--- /dev/null
+++ b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikUserVoiceCreateReqVO.java
@@ -0,0 +1,38 @@
+package cn.iocoder.yudao.module.tik.voice.vo;
+
+import io.swagger.v3.oas.annotations.media.Schema;
+import jakarta.validation.constraints.NotBlank;
+import jakarta.validation.constraints.NotNull;
+import lombok.Data;
+
+/**
+ * 用户 App - 创建配音 Request VO
+ *
+ * @author 芋道源码
+ */
+@Schema(description = "用户 App - 创建配音 Request VO")
+@Data
+public class AppTikUserVoiceCreateReqVO {
+
+    @Schema(description = "配音名称", requiredMode = Schema.RequiredMode.REQUIRED, example = "我的配音")
+    @NotBlank(message = "配音名称不能为空")
+    private String name;
+
+    @Schema(description = "音频文件编号（关联 infra_file.id）", requiredMode = Schema.RequiredMode.REQUIRED, example = "1")
+    @NotNull(message = "音频文件编号不能为空")
+    private Long fileId;
+
+    @Schema(description = "是否自动识别", example = "false")
+    private Boolean autoTranscribe;
+
+    @Schema(description = "语言：zh-CN-简体中文，zh-TW-繁體中文，en-US-English", example = "zh-CN")
+    private String language;
+
+    @Schema(description = "音色类型：female-女声，male-男声", example = "female")
+    private String gender;
+
+    @Schema(description = "备注", example = "这是一个测试配音")
+    private String note;
+
+}
+
diff --git a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikUserVoicePageReqVO.java b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikUserVoicePageReqVO.java
new file mode 100644
index 0000000000..b0939883c1
--- /dev/null
+++ b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikUserVoicePageReqVO.java
@@ -0,0 +1,23 @@
+package cn.iocoder.yudao.module.tik.voice.vo;
+
+import cn.iocoder.yudao.framework.common.pojo.PageParam;
+import io.swagger.v3.oas.annotations.media.Schema;
+import lombok.Data;
+
+/**
+ * 用户 App - 用户配音分页 Request VO
+ *
+ * @author 芋道源码
+ */
+@Schema(description = "用户 App - 用户配音分页 Request VO")
+@Data
+public class AppTikUserVoicePageReqVO extends PageParam {
+
+    @Schema(description = "用户编号（自动填充，无需传递）")
+    private Long userId;
+
+    @Schema(description = "配音名称（模糊查询）", example = "我的配音")
+    private String name;
+
+}
+
diff --git a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikUserVoiceRespVO.java b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikUserVoiceRespVO.java
new file mode 100644
index 0000000000..ff5a1e4993
--- /dev/null
+++ b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikUserVoiceRespVO.java
@@ -0,0 +1,48 @@
+package cn.iocoder.yudao.module.tik.voice.vo;
+
+import io.swagger.v3.oas.annotations.media.Schema;
+import lombok.Data;
+
+import java.time.LocalDateTime;
+
+/**
+ * 用户 App - 用户配音 Response VO
+ *
+ * @author 芋道源码
+ */
+@Schema(description = "用户 App - 用户配音 Response VO")
+@Data
+public class AppTikUserVoiceRespVO {
+
+    @Schema(description = "配音编号", requiredMode = Schema.RequiredMode.REQUIRED, example = "1")
+    private Long id;
+
+    @Schema(description = "配音名称", requiredMode = Schema.RequiredMode.REQUIRED, example = "我的配音")
+    private String name;
+
+    @Schema(description = "音频文件编号（关联 infra_file.id）", requiredMode = Schema.RequiredMode.REQUIRED, example = "1")
+    private Long fileId;
+
+    @Schema(description = "文件访问URL（通过 file_id 关联查询获取）")
+    private String fileUrl;
+
+    @Schema(description = "语音识别内容", example = "这是识别出的文字内容")
+    private String transcription;
+
+    @Schema(description = "语言：zh-CN-简体中文，zh-TW-繁體中文，en-US-English", example = "zh-CN")
+    private String language;
+
+    @Schema(description = "音色类型：female-女声，male-男声", example = "female")
+    private String gender;
+
+    @Schema(description = "备注", example = "这是一个测试配音")
+    private String note;
+
+    @Schema(description = "创建时间", requiredMode = Schema.RequiredMode.REQUIRED)
+    private LocalDateTime createTime;
+
+    @Schema(description = "更新时间", requiredMode = Schema.RequiredMode.REQUIRED)
+    private LocalDateTime updateTime;
+
+}
+
diff --git a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikUserVoiceUpdateReqVO.java b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikUserVoiceUpdateReqVO.java
new file mode 100644
index 0000000000..433a2fb652
--- /dev/null
+++ b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikUserVoiceUpdateReqVO.java
@@ -0,0 +1,36 @@
+package cn.iocoder.yudao.module.tik.voice.vo;
+
+import io.swagger.v3.oas.annotations.media.Schema;
+import jakarta.validation.constraints.NotNull;
+import lombok.Data;
+
+/**
+ * 用户 App - 更新配音 Request VO
+ *
+ * @author 芋道源码
+ */
+@Schema(description = "用户 App - 更新配音 Request VO")
+@Data
+public class AppTikUserVoiceUpdateReqVO {
+
+    @Schema(description = "配音编号", requiredMode = Schema.RequiredMode.REQUIRED, example = "1")
+    @NotNull(message = "配音编号不能为空")
+    private Long id;
+
+    @Schema(description = "配音名称", example = "我的配音")
+    private String name;
+
+    @Schema(description = "语言：zh-CN-简体中文，zh-TW-繁體中文，en-US-English", example = "zh-CN")
+    private String language;
+
+    @Schema(description = "音色类型：female-女声，male-男声", example = "female")
+    private String gender;
+
+    @Schema(description = "备注", example = "这是一个测试配音")
+    private String note;
+
+    @Schema(description = "识别内容", example = "识别文字，可手动编辑")
+    private String transcription;
+
+}
+
diff --git a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikVoicePreviewReqVO.java b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikVoicePreviewReqVO.java
new file mode 100644
index 0000000000..30231e4b74
--- /dev/null
+++ b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikVoicePreviewReqVO.java
@@ -0,0 +1,43 @@
+package cn.iocoder.yudao.module.tik.voice.vo;
+
+import io.swagger.v3.oas.annotations.media.Schema;
+import jakarta.validation.constraints.Size;
+import lombok.Data;
+
+/**
+ * 我的音色试听请求
+ */
+@Data
+public class AppTikVoicePreviewReqVO {
+
+    @Schema(description = "输入文本")
+    @Size(max = 4000, message = "输入文本不能超过 4000 个字符")
+    private String inputText;
+
+    @Schema(description = "识别文本，用于拼接")
+    @Size(max = 4000, message = "识别文本不能超过 4000 个字符")
+    private String transcriptionText;
+
+    @Schema(description = "音色 ID（CosyVoice voiceId）")
+    private String voiceId;
+
+    @Schema(description = "音色源音频 OSS 地址（当没有 voiceId 时必传）")
+    private String fileUrl;
+
+    @Schema(description = "模型名称，默认 cosyvoice-v2")
+    private String model;
+
+    @Schema(description = "语速", example = "1.0")
+    private Float speechRate;
+
+    @Schema(description = "音量", example = "0")
+    private Float volume;
+
+    @Schema(description = "情感", example = "neutral")
+    private String emotion;
+
+    @Schema(description = "音频格式，默认 wav")
+    private String audioFormat;
+}
+
+
diff --git a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikVoicePreviewRespVO.java b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikVoicePreviewRespVO.java
new file mode 100644
index 0000000000..3d3bf18e7f
--- /dev/null
+++ b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikVoicePreviewRespVO.java
@@ -0,0 +1,26 @@
+package cn.iocoder.yudao.module.tik.voice.vo;
+
+import io.swagger.v3.oas.annotations.media.Schema;
+import lombok.Data;
+
+@Data
+@Schema(description = "音色试听响应")
+public class AppTikVoicePreviewRespVO {
+
+    @Schema(description = "音频播放地址（预签名 URL）")
+    private String audioUrl;
+
+    @Schema(description = "音频格式", example = "wav")
+    private String format;
+
+    @Schema(description = "采样率", example = "24000")
+    private Integer sampleRate;
+
+    @Schema(description = "CosyVoice 请求ID")
+    private String requestId;
+
+    @Schema(description = "使用的音色 ID")
+    private String voiceId;
+}
+
+
diff --git a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikVoiceTtsReqVO.java b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikVoiceTtsReqVO.java
new file mode 100644
index 0000000000..5630e18685
--- /dev/null
+++ b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikVoiceTtsReqVO.java
@@ -0,0 +1,46 @@
+package cn.iocoder.yudao.module.tik.voice.vo;
+
+import io.swagger.v3.oas.annotations.media.Schema;
+import jakarta.validation.constraints.Size;
+import lombok.Data;
+
+/**
+ * 文本转语音请求 VO
+ */
+@Data
+public class AppTikVoiceTtsReqVO {
+
+    @Schema(description = "输入文本")
+    @Size(max = 4000, message = "输入文本不能超过 4000 个字符")
+    private String inputText;
+
+    @Schema(description = "识别文本，用于拼接")
+    @Size(max = 4000, message = "识别文本不能超过 4000 个字符")
+    private String transcriptionText;
+
+    @Schema(description = "音色 ID（CosyVoice voiceId）", example = "cosyvoice-v2-myvoice-xxx")
+    private String voiceId;
+
+    @Schema(description = "音色源音频 OSS 地址（当没有 voiceId 时必传）")
+    private String fileUrl;
+
+    @Schema(description = "模型名称，默认 cosyvoice-v2", example = "cosyvoice-v3")
+    private String model;
+
+    @Schema(description = "语速，默认 1.0", example = "1.0")
+    private Float speechRate;
+
+    @Schema(description = "情感", example = "happy")
+    private String emotion;
+
+    @Schema(description = "音量调节范围 [-10,10]", example = "0")
+    private Float volume;
+
+    @Schema(description = "目标采样率，默认 24000")
+    private Integer sampleRate;
+
+    @Schema(description = "音频格式，默认 wav，可选 mp3")
+    private String audioFormat;
+}
+
+
diff --git a/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikVoiceTtsRespVO.java b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikVoiceTtsRespVO.java
new file mode 100644
index 0000000000..0b386389c7
--- /dev/null
+++ b/yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikVoiceTtsRespVO.java
@@ -0,0 +1,29 @@
+package cn.iocoder.yudao.module.tik.voice.vo;
+
+import io.swagger.v3.oas.annotations.media.Schema;
+import lombok.Data;
+
+@Data
+@Schema(description = "CosyVoice 文本转语音响应")
+public class AppTikVoiceTtsRespVO {
+
+    @Schema(description = "用户文件编号", example = "1024")
+    private Long fileId;
+
+    @Schema(description = "音频播放地址（预签名 URL）")
+    private String audioUrl;
+
+    @Schema(description = "音频格式", example = "mp3")
+    private String format;
+
+    @Schema(description = "采样率", example = "24000")
+    private Integer sampleRate;
+
+    @Schema(description = "CosyVoice 请求ID")
+    private String requestId;
+
+    @Schema(description = "使用的音色 ID")
+    private String voiceId;
+}
+
+
diff --git a/yudao-module-tik/src/test/java/cn/iocoder/yudao/module/tik/voice/service/LatentsyncServiceImplTest.java b/yudao-module-tik/src/test/java/cn/iocoder/yudao/module/tik/voice/service/LatentsyncServiceImplTest.java
new file mode 100644
index 0000000000..f2d18fd776
--- /dev/null
+++ b/yudao-module-tik/src/test/java/cn/iocoder/yudao/module/tik/voice/service/LatentsyncServiceImplTest.java
@@ -0,0 +1,62 @@
+package cn.iocoder.yudao.module.tik.voice.service;
+
+import cn.iocoder.yudao.module.tik.voice.client.LatentsyncClient;
+import cn.iocoder.yudao.module.tik.voice.client.dto.LatentsyncSubmitRequest;
+import cn.iocoder.yudao.module.tik.voice.client.dto.LatentsyncSubmitResponse;
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikLatentsyncSubmitReqVO;
+import cn.iocoder.yudao.module.tik.voice.vo.AppTikLatentsyncSubmitRespVO;
+import org.junit.jupiter.api.BeforeEach;
+import org.junit.jupiter.api.Test;
+import org.junit.jupiter.api.extension.ExtendWith;
+import org.mockito.ArgumentCaptor;
+import org.mockito.Mock;
+import org.mockito.junit.jupiter.MockitoExtension;
+
+import static org.assertj.core.api.Assertions.assertThat;
+import static org.mockito.Mockito.verify;
+import static org.mockito.Mockito.when;
+
+@ExtendWith(MockitoExtension.class)
+class LatentsyncServiceImplTest {
+
+    @Mock
+    private LatentsyncClient latentsyncClient;
+
+    private LatentsyncServiceImpl latentsyncService;
+
+    @BeforeEach
+    void setUp() {
+        latentsyncService = new LatentsyncServiceImpl(latentsyncClient);
+    }
+
+    @Test
+    void submitTask_success() {
+        AppTikLatentsyncSubmitReqVO reqVO = new AppTikLatentsyncSubmitReqVO();
+        reqVO.setAudioUrl("https://cdn.example.com/audio.wav");
+        reqVO.setVideoUrl("https://cdn.example.com/video.mp4");
+        reqVO.setGuidanceScale(2);
+        reqVO.setSeed(999);
+
+        LatentsyncSubmitResponse clientResp = new LatentsyncSubmitResponse();
+        clientResp.setRequestId("task-123");
+        clientResp.setStatus("IN_QUEUE");
+        clientResp.setQueuePosition(0);
+        when(latentsyncClient.submitTask(org.mockito.Mockito.any())).thenReturn(clientResp);
+
+        AppTikLatentsyncSubmitRespVO respVO = latentsyncService.submitTask(reqVO);
+
+        assertThat(respVO.getRequestId()).isEqualTo("task-123");
+        assertThat(respVO.getStatus()).isEqualTo("IN_QUEUE");
+        assertThat(respVO.getQueuePosition()).isZero();
+
+        ArgumentCaptor<LatentsyncSubmitRequest> captor = ArgumentCaptor.forClass(LatentsyncSubmitRequest.class);
+        verify(latentsyncClient).submitTask(captor.capture());
+        LatentsyncSubmitRequest submitRequest = captor.getValue();
+        assertThat(submitRequest.getAudioUrl()).isEqualTo(reqVO.getAudioUrl());
+        assertThat(submitRequest.getVideoUrl()).isEqualTo(reqVO.getVideoUrl());
+        assertThat(submitRequest.getGuidanceScale()).isEqualTo(reqVO.getGuidanceScale());
+        assertThat(submitRequest.getSeed()).isEqualTo(reqVO.getSeed());
+    }
+}
+
+
diff --git a/yudao-server/src/main/resources/application-local.yaml b/yudao-server/src/main/resources/application-local.yaml
index dd438bfb55..3c0fc150a9 100644
--- a/yudao-server/src/main/resources/application-local.yaml
+++ b/yudao-server/src/main/resources/application-local.yaml
@@ -223,6 +223,11 @@ wx:
 
 # 芋道配置项，设置当前项目所有自定义的配置
 yudao:
+  cosyvoice:
+    api-key: sk-10c746f8cb8640738f8d6b71af699003
+  # tik:
+  #   latentsync:
+  #     api-key: ${TIK_LATENTSYNC_API_KEY:} # 建议通过环境变量覆盖仓库默认值
   captcha:
     enable: false # 本地环境，暂时关闭图片验证码，方便登录等接口的测试；
   security:
@@ -265,4 +270,4 @@ justauth:
   cache:
     type: REDIS
     prefix: 'social_auth_state:' # 缓存前缀，目前只对 Redis 缓存生效，默认 JUSTAUTH::STATE::
-    timeout: 24h # 超时时长，目前只对 Redis 缓存生效，默认 3 分钟
\ No newline at end of file
+    timeout: 24h # 超时时长，目前只对 Redis 缓存生效，默认 3 分钟
diff --git a/yudao-server/src/main/resources/application.yaml b/yudao-server/src/main/resources/application.yaml
index 98315b3424..ce3848874e 100644
--- a/yudao-server/src/main/resources/application.yaml
+++ b/yudao-server/src/main/resources/application.yaml
@@ -213,6 +213,13 @@ spring:
               sse-endpoint: /sse
 
 yudao:
+  cosyvoice:
+    enabled: true
+    api-key: sk-10c746f8cb8640738f8d6b71af699003
+    default-model: cosyvoice-v2
+    sample-rate: 24000
+    audio-format: mp3
+    preview-text: 您好，欢迎体验专属音色
   ai:
     gemini: # 谷歌 Gemini
       enable: true