fix:问题

.claude/settings.local.json

@@ -82,7 +82,8 @@
       "Skill(pptx:*)",
       "Bash(pdftoppm:*)",
       "Bash(pip install:*)",
-      "Bash(where:*)"
+      "Bash(where:*)",
+      "mcp__web-reader__webReader"
     ],
     "deny": [],
     "ask": []

frontend/app/web-gold/src/api/voice.js

@@ -22,6 +22,7 @@ export const VoiceService = {
    * @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）
+   * @param {string} data.text - 音频文本（用于语音复刻，前端通过音频识别获取）
    * @param {string} data.providerType - 供应商类型（可选）：cosyvoice-阿里云，siliconflow-硅基流动
    * @returns {Promise}
    */

frontend/app/web-gold/src/composables/useUpload.js

@@ -133,28 +133,23 @@ export function useUpload() {
         fileType: file.type,
         groupId,
         coverBase64,
-        duration: file.type.startsWith('video/') ? null : undefined // 视频时长由后端处理或前端可选传递
+        duration: file.type.startsWith('video/') ? null : undefined
       })
 
-      // 设置成功状态
       state.uploading = false
       state.status = 'success'
       state.progress = 100
 
-      // 通知成功
       const fileId = completeData.data?.infraFileId || completeData.data?.userFileId
-      onSuccess && onSuccess(fileId)
+      const fileUrl = presignedData.data.presignedUrl
+      onSuccess && onSuccess(fileId, fileUrl)
 
       return fileId
     } catch (error) {
-      // 设置错误状态
       state.uploading = false
       state.status = 'error'
       state.error = error.message || '上传失败'
-
-      // 通知错误
       onError && onError(error)
-
       throw error
     }
   }

frontend/app/web-gold/src/views/dh/VoiceCopy.vue

@@ -111,6 +111,7 @@ import { PlusOutlined, SearchOutlined, UploadOutlined, PlayCircleOutlined } from
 import { VoiceService } from '@/api/voice'
 import { MaterialService } from '@/api/material'
 import { useUpload } from '@/composables/useUpload'
+import useVoiceText from '@gold/hooks/web/useVoiceText'
 import dayjs from 'dayjs'
 import BasicLayout from '@/layouts/components/BasicLayout.vue'
 
@@ -123,7 +124,9 @@ const DEFAULT_FORM_DATA = {
   autoTranscribe: true,
   language: 'zh-CN',
   gender: 'female',
-  note: ''
+  note: '',
+  text: '', // 音频文本
+  fileUrl: '' // 文件URL（用于获取音频文本）
 }
 
 // ========== 响应式数据 ==========
@@ -155,6 +158,9 @@ const formData = reactive({ ...DEFAULT_FORM_DATA })
 // ========== Upload Hook ==========
 const { state: uploadState, upload } = useUpload()
 
+// ========== VoiceText Hook ==========
+const { getVoiceText } = useVoiceText()
+
 // ========== 计算属性 ==========
 const isCreateMode = computed(() => formMode.value === 'create')
 
@@ -307,13 +313,16 @@ const handleCustomUpload = async (options) => {
   try {
     const fileId = await upload(file, {
       fileCategory: 'voice',
-      groupId: null, // 配音模块不使用groupId
+      groupId: null,
       coverBase64: null,
       onStart: () => {},
       onProgress: () => {},
-      onSuccess: (id) => {
+      onSuccess: async (id, fileUrl) => {
         formData.fileId = id
+        formData.fileUrl = fileUrl // 保存文件URL
         message.success('文件上传成功')
+        // 通过fileId获取播放URL用于语音识别
+        await fetchAudioTextById(id)
         onSuccess?.({ code: 0, data: id }, file)
       },
       onError: (error) => {
@@ -330,12 +339,51 @@ const handleCustomUpload = async (options) => {
   }
 }
 
+
+
+// 通过fileId获取音频文本
+const fetchAudioTextById = async (fileId) => {
+  if (!fileId) return
+  try {
+    // 获取音频播放URL
+    const res = await MaterialService.getAudioPlayUrl(fileId)
+    if (res.code === 0 && res.data) {
+      const rawFileUrl = res.data
+      const results = await getVoiceText([{ audio_url: rawFileUrl }])
+      if (results && results.length > 0) {
+        const text = results[0].value
+        formData.text = text
+        if (text) {
+          message.success('音频文本获取成功')
+        }
+      }
+    }
+  } catch (error) {
+    console.error('获取音频文本失败:', error)
+  }
+}
+
+// 获取音频文本
+const fetchAudioText = async (fileUrl) => {
+  if (!fileUrl) return
+  try {
+    // 阿里云语音识别服务无法访问预签名URL，使用原始URL
+    const rawFileUrl = extractRawUrl(fileUrl)
+    const results = await getVoiceText([{ audio_url: rawFileUrl }])
+    if (results && results.length > 0) {
+      const text = results[0].value
+      formData.text = text
+      if (text) {
+        message.success('音频文本获取成功')
+      }
+    }
+  } catch (error) {
+    console.error('获取音频文本失败:', error)
+  }
+}
+
 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')
   }
@@ -363,7 +411,8 @@ const handleSubmit = async () => {
         autoTranscribe: formData.autoTranscribe,
         language: formData.language,
         gender: formData.gender,
-        note: formData.note
+        note: formData.note,
+        text: formData.text // 传入音频文本
       }
     : {
         id: formData.id,

yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/client/SiliconFlowProvider.java

@@ -58,7 +58,6 @@ public class SiliconFlowProvider implements VoiceCloneProvider {
             sfRequest.setText(getOrDefault(request.getTranscriptionText(), config.getPreviewText()));
             sfRequest.setAudio(AUDIO_MIME_TYPE + base64Audio);
 
-            // 调用上传参考音频 API
             String url = config.getBaseUrl() + config.getVoiceUploadUrl();
             String requestBody = JSONUtil.toJsonStr(sfRequest);
             log.debug("[SiliconFlowProvider][请求体]{}", requestBody);
@@ -123,7 +122,6 @@ public class SiliconFlowProvider implements VoiceCloneProvider {
                     .responseFormat(getOrDefault(request.getAudioFormat(), config.getAudioFormat()))
                     .build();
 
-            // 调用文本转语音 API
             String url = config.getBaseUrl() + config.getTtsUrl();
             String requestBody = JSONUtil.toJsonStr(sfRequest);
             log.debug("[SiliconFlowProvider][请求体]{}", requestBody);
@@ -142,7 +140,6 @@ public class SiliconFlowProvider implements VoiceCloneProvider {
                 throw new RuntimeException("硅基流动文本转语音失败: " + errorBody);
             }
 
-            // 硅基流动直接返回二进制音频数据
             byte[] audioBytes = response.bodyBytes();
             String base64Audio = Base64.getEncoder().encodeToString(audioBytes);
 

yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/client/dto/SiliconFlowTtsRequest.java

@@ -36,8 +36,6 @@ public class SiliconFlowTtsRequest {
      */
     private Float speed;
 
-
-
     /**
      * 响应格式（mp3, opus, wav, pcm）（API 参数名：response_format）
      */

yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/config/SiliconFlowProviderConfig.java

@@ -32,8 +32,11 @@ public class SiliconFlowProviderConfig extends VoiceProviderProperties.ProviderC
 
     /**
      * 默认采样率
+     * <p>mp3: 32000, 44100 (默认 44100)</p>
+     * <p>opus: 48000</p>
+     * <p>wav/pcm: 8000, 16000, 24000, 32000, 44100 (默认 44100)</p>
      */
-    private Integer sampleRate = 24000;
+    private Integer sampleRate = 44100;
 
     /**
      * 默认音频格式

yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/service/TikUserVoiceServiceImpl.java

@@ -144,34 +144,38 @@ public class TikUserVoiceServiceImpl implements TikUserVoiceService {
                 .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
+                .setTranscription(createReqVO.getText()); // 使用前端传入的文本
         voiceMapper.insert(voice);
 
         // 4. 调用语音克隆服务，生成 voice_id
-        try {
+        if (StrUtil.isNotBlank(createReqVO.getText())) {
-            log.info("[createVoice][开始语音复刻，配音编号({})，文件ID({})，供应商({})]",
+            try {
-                    voice.getId(), fileDO.getId(), createReqVO.getProviderType());
+                log.info("[createVoice][开始语音复刻，配音编号({})，文件ID({})，供应商({})]",
-            String fileAccessUrl = fileApi.presignGetUrl(fileDO.getUrl(), PRESIGN_URL_EXPIRATION_SECONDS);
+                        voice.getId(), fileDO.getId(), createReqVO.getProviderType());
+                String fileAccessUrl = fileApi.presignGetUrl(fileDO.getUrl(), PRESIGN_URL_EXPIRATION_SECONDS);
 
-            VoiceCloneProvider provider = voiceProviderFactory.getProvider(createReqVO.getProviderType());
+                VoiceCloneProvider provider = voiceProviderFactory.getProvider(createReqVO.getProviderType());
-            String providerType = getProviderType(createReqVO.getProviderType(), provider);
+                String providerType = getProviderType(createReqVO.getProviderType(), provider);
-            String model = getModelByProvider(providerType);
+                String model = getModelByProvider(providerType);
 
-            VoiceCloneRequest cloneRequest = new VoiceCloneRequest();
+                VoiceCloneRequest cloneRequest = new VoiceCloneRequest();
-            cloneRequest.setAudioUrl(fileAccessUrl);
+                cloneRequest.setAudioUrl(fileAccessUrl);
-            cloneRequest.setModel(model);
+                cloneRequest.setModel(model);
-            cloneRequest.setPrefix("voice" + voice.getId());
+                cloneRequest.setPrefix("voice" + voice.getId());
-            cloneRequest.setTranscriptionText(voice.getTranscription());
+                cloneRequest.setTranscriptionText(createReqVO.getText()); // 使用前端传入的文本
 
-            VoiceCloneResult cloneResult = provider.cloneVoice(cloneRequest);
+                VoiceCloneResult cloneResult = provider.cloneVoice(cloneRequest);
-            String voiceId = cloneResult.getVoiceId();
+                String voiceId = cloneResult.getVoiceId();
 
-            voice.setVoiceId(voiceId);
+                voice.setVoiceId(voiceId);
-            voiceMapper.updateById(voice);
+                voiceMapper.updateById(voice);
 
-            log.info("[createVoice][语音复刻成功，配音编号({})，voice_id({})]", voice.getId(), voiceId);
+                log.info("[createVoice][语音复刻成功，配音编号({})，voice_id({})]", voice.getId(), voiceId);
-        } catch (Exception e) {
+            } catch (Exception e) {
-            log.error("[createVoice][语音复刻失败，配音编号({})，错误信息: {}]", voice.getId(), e.getMessage(), e);
+                log.error("[createVoice][语音复刻失败，配音编号({})，错误信息: {}]", voice.getId(), e.getMessage(), e);
+            }
+        } else {
+            log.info("[createVoice][未提供文本，跳过语音复刻，配音编号({})]", voice.getId());
         }
 
 

yudao-module-tik/src/main/java/cn/iocoder/yudao/module/tik/voice/vo/AppTikUserVoiceCreateReqVO.java

@@ -3,6 +3,7 @@ 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 jakarta.validation.constraints.Size;
 import lombok.Data;
 
 /**
@@ -34,6 +35,10 @@ public class AppTikUserVoiceCreateReqVO {
     @Schema(description = "备注", example = "这是一个测试配音")
     private String note;
 
+    @Schema(description = "音频文本（用于语音复刻，前端通过音频识别获取）")
+    @Size(max = 4000, message = "音频文本不能超过 4000 个字符")
+    private String text;
+
     @Schema(description = "供应商类型：cosyvoice-阿里云，siliconflow-硅基流动（不传则使用默认）", example = "cosyvoice")
     private String providerType;