-
#8753 Audio/Video Speed Sync Stage An Error Occurred Decoding failed. ffmpeg returned error code: 3199971767...
💬 1[Last reply:3 days ago] The direct cause: the dubbing fragment dubb-1-7586748a8100d6a7533f3d8e205c8d38.wav generated by Edge-TTS is an invalid/empty audio file (0 bytes or incomplete data), so when the alignment stage calls ...91.196**4 Posted at: 3 days ago -
#8751 Audio/Video Speed Sync Stage An Error Occurred Decoding failed. ffmpeg returned error code: 3221225477...
💬 1[Last reply:3 days ago] The ffmpeg return code 3221225477 is the decimal form of 0xC0000005 (memory access violation), meaning the ffmpeg process crashed while decoding one of the dubbed audio segments during the A/V sync st...91.196**4 Posted at: 3 days ago -
#8750 Audio/Video Speed Sync Stage An Error Occurred Expecting value: line 1 column 1 (char 0)Traceback (most recent call last):...
💬 1[Last reply:3 days ago] The direct cause: in the audio/video alignment stage, pydub calls ffprobe to read a dubbing clip, and ffprobe returned an empty result. This is because the audio file being read does not exist, is 0 b...91.196**4 Posted at: 3 days ago -
#8749 Audio/Video Speed Sync Stage An Error Occurred Decoding failed. ffmpeg returned error code: 3199971767...
💬 1[Last reply:3 days ago] 核心原因：第15条字幕生成的配音文件 dubb-15-....wav 是空文件或损坏文件（大小为0或内容不是有效音频），pydub 调用 ffmpeg 解码时因此报 Invalid data found when processing input。这通常表示 Edge-TTS 对该行没有返回有效音频，或者该行字幕文本为空、内容异常，也可能是对短片段进行静音裁剪后变成了0长度。处理步骤：清理失败缓存...91.196**4 Posted at: 3 days ago -
#8748 配音全部失败 :403, message='Invalid response status', url='wss://speech.platform.bing.com/consumer/speech/synthesize/readaloud/edge/v1?TrustedClientToken=6A5AA1D4EAFF4E9FB37E23D68491D6F4...
💬 1[Last reply:3 days ago] 这是 Edge-TTS 微软免费接口的验证逻辑发生变更导致的 403 拒绝（并非网络代理问题），你当前使用的 v4.08 版本过旧，内置的 edge-tts 库已无法通过微软校验。解决方案（任选其一）：升级软件（推荐，根治）前往官网下载最新完整包或补丁包覆盖：https://pyvideotrans.com/downpackage覆盖时必须把补丁包中的 所有文件和文件夹 一并覆盖，尤其是 sp.e...185.244**7 Posted at: 3 days ago -
#8747 Video file selection required...
💬 1[Last reply:3 days ago] The selected file was not recognized as a valid video path. The usual causes are: a long file name with emoji/special characters, a path that is too deep or contains special symbols, a file whose exte...111.125**5 Posted at: 3 days ago -
#8746 语音识别阶段出错[faster-whisper(内置)] Unable to open file 'model.bin' in model 'C:/Users/yg1/aaaaa/models/models--distil-whisper--distil-large-v3.5-ct2',Traceback (most recent call last): ...
💬 1[Last reply:3 days ago] 这是因为该模型文件未完整下载或下载失败：model.bin 不存在，无法加载。另外，distil-large-v3.5 是为英文优化的蒸馏模型，仅当发音语言是英语时适用；当前发音语言是韩语，不适合使用该模型。解决方案1. 改用正确的模型（推荐，最省事）在语音识别渠道处，将模型改为 large-v3-turbo（推荐）或 large-v3，然后重新执行。韩语等非英语场景必须使用这两个通用模型。2. ...216.151**2 Posted at: 3 days ago -
#8745 Error during merge stage Video translation complete, but failed to move output file. Copy manually from:D:/win-pyvideotrans-v4.18/tmp/21420/43c16208cb/laste_target.mp4...
💬 1[Last reply:3 days ago] The actual cause is that the full target path exceeds Windows' 260-character limit (MAX_PATH). In your case the video name alone is about 100 characters, so the string C:\Users\Son\Downloads\xbd-downl...193.176**9 Posted at: 3 days ago -
#8743 403, message='Invalid response status', url='wss://speech.platform.bing.com/consumer/speech/synthesize/readaloud/edge/v1?TrustedClientToken=6A5AA1D4EAFF4E9FB37E23D68491D6F4&Sec-MS-...
💬 1[Last reply:3 days ago] 这是 Edge-TTS 免费接口的验证逻辑失效导致的。微软已更换鉴权方式，旧版本内置的 edge-tts 请求会被服务端直接拒绝，返回 403。你的版本是 v3.79，该问题在 v4.16 才修复。解决方案：到官网下载最新的 完整包（约 2.6GB+，不要只下 260MB 的补丁包）：https://pyvideotrans.com/downpackage解压到简短的全英文路径（如 D:\pyVi...212.135**8 Posted at: 3 days ago -
#8742 kokoro(LocalAPI): [Kokoro-TTS] Service must be deployed and running before useself.api_url='http://127.0.0.1:5066/v1/audio/speech'...
💬 1[Last reply:3 days ago] Core cause: the local Kokoro-TTS service is not running (or is not listening on port 5066). The error WinError 10061 ConnectionRefusedError / Max retries exceeded ... /v1/audio/speech means nothing ac...2a01**2211 Posted at: 3 days ago -
#8741 翻译字幕阶段出错 [Google(免费)] Secure connection failed, check system time or network settingsurllib3.exceptions.SSLError: EOF occurred in violation of protocol (_ssl.c:1017)...
💬 1[Last reply:4 days ago] 这是网络层 TLS 连接被中断导致的：Google 免费翻译在中国大陆无法直连，而回退的微软免费翻译（edge.microsoft.com）又在使用代理时被中断，说明你填写的代理不可用，或未填写代理，或系统时间异常导致 SSL 校验失败。解决步骤（按顺序排查）：检查主界面代理设置：在“对齐控制”行右侧的“网络代理”文本框中确认代理地址格式正确（如 http://127.0.0.1:7890），并确...188.253**2 Posted at: 4 days ago -
#8740 Error during subtitle translation stage [M2M100(Built-in)] System Error (None): NoneTraceback (most recent call last):...
💬 1[Last reply:4 days ago] CauseThe M2M100 local model is incomplete or was not downloaded — the file models/m2m100_12b/sentencepiece.model does not exist, so loading the tokenizer fails directly with OSError: Not found. This i...212.180**4 Posted at: 4 days ago -
#8738 Proxy configuration issue, check settings or disable proxyTraceback (most recent call last):...
💬 1[Last reply:4 days ago] 这是代理地址填写错误或代理服务未启动导致的：日志中记录了 代理地址: http://127.0.0.1:4088，但该端口上并没有代理服务在监听（WinError 10061 目标计算机积极拒绝），导致所有网络请求（包括微软翻译的 edge.microsoft.com）都经由此代理发出并被拒绝。解决方案（按需二选一）：如果你没有运行代理软件（最常见情况）将主界面对齐行（字幕、配音、画面对齐那一行）...120.238**8 Posted at: 4 days ago -
#8737 OpenRouter: anthropic/claude-sonnet-4 is no longer available on OpenRouter: no provider serves this model anymore. See available models at https://openrouter.ai/modelshttps://openr...
💬 1[Last reply:4 days ago] 这不是软件本身的 bug，而是所选的 OpenRouter 模型 anthropic/claude-sonnet-4 已被下线：OpenRouter 服务端已没有任何 provider 提供该模型，接口返回 404 no provider serves this model anymore，因此翻译任务直接中断。解决步骤：先访问 https://openrouter.ai/models 查看当前可...156.229**7 Posted at: 4 days ago -
#8736 语音识别阶段出错[Qwen-ASR(内置)] CUDA out of memory. Tried to allocate 15.83 GiB. GPU 0 has a total capacity of 8.00 GiB of which 0 bytes is free. Of the allocated memory 5.17 GiB is allocat...
💬 1[Last reply:4 days ago] 核心原因是显卡显存物理不足：你的 GPU 0 只有 8 GB 显存，当前已被 PyTorch 占用约 5.17 GB，而本次请求需一次性再分配 15.83 GiB，因此必然 OOM 失败。Qwen-ASR 在 GPU 上运行对切片长度和显存很敏感，同时保留原模型占用，导致 8 GB 不够用。解决方案（按优先级）取消勾选 CUDA加速，改用 CPU 运行 Qwen-ASR 0.6B（最稳妥）0.6B...14.212**1 Posted at: 4 days ago -
#8735 中文电影那个模型最准...
💬 1[Last reply:4 days ago] 中文视频识别准确度最高的是收费在线渠道（豆包语音识别大模型极速版、小米 mimo-v2.5-asr）；在免费且本地可用的渠道里，首选 Qwen-ASR(内置) 1.7B。电影通常背景音乐和音效较重，只更换模型往往不够；还必须配合人声分离和 LLM 纠错。推荐组合（按优先级）收费渠道（最准）：选择 豆包语音识别大模型极速版；在 菜单-翻译设置 中填写 API Key，然后在主界面语音识别渠道中选择该...120.217**0 Posted at: 4 days ago -
#8734 Dubbing failed completely:403, message='Invalid response status', url='wss://speech.platform.bing.com/consumer/speech/synthesize/readaloud/edge/v1?TrustedClientToken=6A5AA1D4EAFF4E...
💬 1[Last reply:4 days ago] This is because Microsoft changed the authentication logic of its free edge-tts interface, causing 403 rejections. This is a known issue in v4.15; it is not a problem with your configuration or networ...2405**b4d6 Posted at: 4 days ago -
#8733 Invalid source subtitle interval: 201800-204600Traceback (most recent call last):...
💬 1[Last reply:4 days ago] 这是音画对齐阶段（shot_dubbing_alignment.py）对源字幕时间轴做合法性校验时触发的断言失败：报错中的 201800-204600 是以毫秒表示的区间（201.8s–204.6s，时长 2.8s），该区间落在视频总时长 256.4s 之内，因此不是“超出视频长度”，而是该条字幕在内部数据中的起止时间被判为无效，说明上游识别或翻译产出的字幕时间轴已经错乱，例如起止时间倒置、与相邻...198.65**3 Posted at: 4 days ago -
#8732 Invalid source subtitle interval: 201800-204600Traceback (most recent call last):...
💬 1[Last reply:4 days ago] 报错本质是：对齐引擎 align_shot_dubbing 在校验待对齐的字幕列表时，遇到一条“无效的源字幕时间区间” 201800-204600（即 201.8 秒 ~ 204.6 秒），校验未通过就抛出异常中断，与显存、网络、翻译接口均无关。这条字幕被判定“无效”，常见原因是该行的时间码与相邻行发生重叠、倒序或跨度为 0/异常（例如上一行的结束时间晚于 201.8 秒，或该行结束时间被前面的行...198.65**3 Posted at: 4 days ago -
#8731 Invalid source subtitle interval: 201800-204600Traceback (most recent call last):...
💬 1[Last reply:4 days ago] 这是音画对齐阶段（shot_dubbing_alignment.py）对源字幕时间轴做校验时失败：某条字幕的 开始-结束 区间被判为无效（201800-204600，即 201.8s → 204.6s）。该区间本身起止顺序正常，且未超出视频总时长（256.4s），说明是字幕之间存在重叠、越界，或该片段在裁剪/对齐后时间基准错位，导致校验不通过。知识库中没有收录这条具体报错的成因，无法确定是哪一个环...198.65**3 Posted at: 4 days ago
Open source and free maintenance is not easy. If this project is helpful to you, please consider making a small donation to help the project continue to maintain and update.