#8285 Dubbing failed{'code': 1, 'msg':

85.232**1 Posted at: 1 hour ago

Dubbing failed{'code': 1, 'msg': "isin_mps_friendly() got an unexpected keyword argument 'elements'"}
self.model_name=''
self=[CloneVoice]: uuid='3fc40ef5bb', proxy_str=None, tts_type=39, len=36, language='ru', play=False, volume='+100%', pitch='+0Hz', rate='+0%', speed=1.0, has_done=0, wait_sec=1.0, dub_nums=1, error='{\'code\': 1, \'msg\': "isin_mps_friendly() got an unexpected keyword argument \'elements\'"}', api_url='http://127.0.0.1:9988', is_cuda=True, local_dir=None, is_redubb=False, model_name='', ainame=None
Traceback (most recent call last):
File "videotrans\task\only_one.py", line 108, in run
File "videotrans\task\_stage_dubbing.py", line 24, in dubbing
File "videotrans\task\_stage_dubbing.py", line 101, in _tts
File "videotrans\tts\__init__.py", line 132, in run
File "videotrans\tts\_base.py", line 140, in run
videotrans.configure.excepts.DubbingSrtError: Dubbing failed{'code': 1, 'msg': "isin_mps_friendly() got an unexpected keyword argument 'elements'"}
self.model_name=''
self=[CloneVoice]: uuid='3fc40ef5bb', proxy_str=None, tts_type=39, len=36, language='ru', play=False, volume='+100%', pitch='+0Hz', rate='+0%', speed=1.0, has_done=0, wait_sec=1.0, dub_nums=1, error='{\'code\': 1, \'msg\': "isin_mps_friendly() got an unexpected keyword argument \'elements\'"}', api_url='http://127.0.0.1:9988', is_cuda=True, local_dir=None, is_redubb=False, model_name='', ainame=None

trk=[TransCreate]: uuid='3fc40ef5bb', proxy_str=None, cfg=[TaskCfgVTT]当前工作模式: 翻译视频 单视频模式
原始输入文件名: D:/2_УРОКИ ПО FotoFusion/00_НЕЙРОСЕТЬ/000_Capture One Pro/000_Курс «Захвати одного эксперта»/videos/Практика редактирования из.mp4,
输出结果保存到文件夹: D:/2_УРОКИ ПО FotoFusion/00_НЕЙРОСЕТЬ/000_Capture One Pro/000_Курс «Захвати одного эксперта»/Lesson RU/Практика редактирования из-mp4,
临时文件夹: D:/pyVideoTrans/tmp/12848/3fc40ef5bb
已选 清理已存在
已启用CUDA加速
未选 降噪
已选 识别说话人,最大说话人数量不限制
Сервис STT (распознавание):faster-whisper(Built-in), large-v3-turbo, 发音语言: en, LLM纠错
Канал перевода:DeepSeek,原始语言:en,目标语言:ru, 已选 Отправить полный SRT
Канал озвучки (TTS):clone-voice(LocalAPI), 角色:clone, 配音语言:ru, 未选 二次语音识别
音量:+100%, 语速:+0%, 已选 音频加速, 已选 视频慢速
已选 移除每条字幕配音开头和结尾静音缓冲
静音移除力度: default
字幕: nosubtitle , precent=18, hasend=False, should_recogn=True, should_trans=True, should_dubbing=True, should_hebing=True, is_copy_video=True, video_codec_num=264, ignore_align=False, is_audio_trans=False, clone_ref='', should_recogn2=False, do_diarize=True
video_duration_ms=231660


=
system:Windows-10-10.0.19045-SP0
version:v4.16
frozen:True
language:ru
root_dir:D:/pyVideoTrans

1 Replies

#1 DeepSeek reply at 1 hour ago

This error does not come from pyVideoTrans itself — it is returned by the external clone-voice local API service at http://127.0.0.1:9988 (note error='{"code":1,"msg":"isin_mps_friendly() got an unexpected keyword argument \'elements\'"}', i.e. the service replied with its own error payload). pyVideoTrans only forwarded it and wrapped it as DubbingSrtError. The message is a Python library version conflict inside the clone-voice environment (incompatible torch/pandas/transformers versions in that project's venv), not a settings problem in pyVideoTrans.

The clone-voice channel has been unmaintained for a long time (rated ⭐ in the docs), so the most reliable fix is to switch channels.

Solution (pick one):

  1. Recommended — switch away from clone-voice (keeps clone voice cloning working):

    • OmniVoice-TTS (built-in) — supports all built-in languages incl. ru, built-in, best coverage.
    • Qwen3-TTS (built-in) — supports ru, built-in.
    • Confucius-TTS (built-in) — supports ru.
    • F5-TTS (built-in) — has a dedicated Russian model.
    • Or a maintained local API: GPT-SoVITS / Index-TTS / VoxCPM-TTS / audio.cpp.
      Change the "TTS/配音渠道" on the main window to one of the above and keep 配音角色 = clone.
  2. If you must keep clone-voice: rebuild its Python environment with mutually compatible versions (the isin_mps_friendly signature mismatch means a torch/pandas/transformers API change). Since the project is no longer maintained, pin an older working combination in its own venv, or check its GitHub issues for the matching version set — this is outside pyVideoTrans and cannot be fixed from the client side.
  3. Clone quality checklist (applies to any clone channel, your current config violates parts of it) — 菜单 → 工具 → 高级选项 → 语音识别参数:

    • 最长语音持续(秒) = 6–10
    • 最短语音持续(毫秒) = 3000–4000 (below 3000 ms the reference clip is too short, causing failure/distortion/noise)
    • In 设置更多参数 enable 分离人声背景声 to improve reference audio quality.

Note for your current task: source speech is en, target is ru; the ru clone will still need one of the channels in step 1 since clone-voice is the failing component.

请查阅相关文档:

Thinking...
Post Your Reply
If you have more than 50 messages, you will no longer receive an automatic reply. Please ask a new question!
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.

Dondate via ko-fi.com

Related resource