feat(video-pipeline): 用 ffprobe 实际测量音视频时长并统一字段名为 script

将项目中的 `narration` 字段统一重命名为 `script`,并新增 `getAudioDurationSec` 函数通过 `ffprobe` 实际测量音频和视频文件的时长,替代 Manifest 中的估计值,提高时间线组装的准确性。同时优化字幕逻辑,仅在有 TTS 音频时调整视频速度。
This commit is contained in:
2026-05-01 01:52:02 +08:00
parent 7d526d2b60
commit c878abc39b
8 changed files with 130 additions and 41 deletions

View File

@@ -40,7 +40,7 @@ function initManifest(options) {
}
// 校验必填字段
const requiredFields = ['shotDesc', 'narration', 'imagePrompt']
const requiredFields = ['shotDesc', 'script', 'imagePrompt']
const resolvedMode = mode || 'single'
for (let i = 0; i < rawItems.length; i++) {
@@ -72,7 +72,7 @@ function initManifest(options) {
id: i + 1,
status: 'pending',
shotDesc: raw.shotDesc || '',
narration: raw.narration || raw.text || '',
script: raw.script || '',
duration: raw.duration || 5,
imagePrompt: raw.imagePrompt,
confirmed: false,

View File

@@ -32,7 +32,7 @@ function validateManifest(manifestPath) {
if (manifest.items && Array.isArray(manifest.items)) {
manifest.items.forEach((item, i) => {
const prefix = `items[${i}]`
if (!item.narration && !item.text) issues.push(`${prefix} 缺少 narration 或 text中文旁白`)
if (!item.script && !item.text) issues.push(`${prefix} 缺少 script 或 text中文文案`)
if (!item.shotDesc) issues.push(`${prefix} 缺少 shotDesc分镜描述`)
if (!item.imagePrompt) issues.push(`${prefix} 缺少 imagePrompt`)
if (manifest.mode === 'framePair' && !item.lastFramePrompt) {

View File

@@ -110,7 +110,7 @@ async function generateGemini(item, idx, dir, imagesDir, ratio, refs) {
const file = (result.savedFiles && result.savedFiles.length > 0)
? renameGeneratedFile(
path.relative(dir, result.savedFiles[0]).replace(/\\/g, '/'),
dir, idx, item.narration || item.shotDesc, ''
dir, idx, item.script || item.shotDesc, ''
)
: null
return { file }
@@ -184,7 +184,7 @@ async function generateKling(item, idx, dir, imagesDir, ratio, refs) {
const file = (result.savedFiles && result.savedFiles.length > 0)
? renameGeneratedFile(
path.relative(dir, result.savedFiles[0]).replace(/\\/g, '/'),
dir, idx, item.narration || item.shotDesc, ''
dir, idx, item.script || item.shotDesc, ''
)
: null
return { file }

View File

@@ -15,7 +15,7 @@ async function phaseTts(manifest, manifestPath, options = {}) {
const { synthesize } = require('../qwen-tts')
const items = manifest.items.filter(it =>
it.status === 'done' && (it.narration || it.text) && !it.audio
it.status === 'done' && (it.script || it.text) && !it.audio
)
if (items.length === 0) { log('tts', '无待处理 item跳过'); return }
@@ -25,7 +25,7 @@ async function phaseTts(manifest, manifestPath, options = {}) {
const item = items[i]
const idx = i + 1
try {
const { filePath, duration } = await synthesize(item.narration || item.text, {
const { filePath, duration } = await synthesize(item.script || item.text, {
outputDir: audioDir,
id: item.id || idx,
voice: manifest.ttsVoice || undefined,
@@ -33,7 +33,7 @@ async function phaseTts(manifest, manifestPath, options = {}) {
})
item.audio = path.relative(dir, filePath).replace(/\\/g, '/')
item.audioDuration = Math.round(duration * 1000) / 1000
log('tts', `[${idx}/${items.length}] ${duration.toFixed(1)}s: ${(item.narration || item.text).substring(0, 30)}...`)
log('tts', `[${idx}/${items.length}] ${duration.toFixed(1)}s: ${(item.script || item.text).substring(0, 30)}...`)
} catch (err) {
item.status = 'failed'
item.error = `TTS失败: ${err.message}`