#4372 TaskCfgVTT(is_cuda=True, uuid='617e6c9a85', cache_folder='D:/win-pyvideotrans-v3.99-428/tmp/2152/617e6c9a85', target_dir

124.77* Posted at: 3 hours ago 👁9

语音识别阶段出错 [faster-whisper(本地)] Traceback (most recent call last):
File "videotrans\process\stt_fun.py", line 347, in faster_whisper
File "faster_whisper\transcribe.py", line 689, in init
RuntimeError: CUDA failed with error CUDA-capable device(s) is/are busy or unavailable

Traceback (most recent call last):
File "videotrans\task\job.py", line 105, in run
File "videotrans\task\trans_create.py", line 380, in recogn
File "videotrans\recognition\__init__.py", line 293, in run
File "videotrans\recognition\_base.py", line 143, in run
File "videotrans\recognition\_overall.py", line 33, in _exec
File "videotrans\recognition\_overall.py", line 105, in _faster
File "videotrans\configure\_base.py", line 289, in _new_process
RuntimeError: Traceback (most recent call last):
File "videotrans\process\stt_fun.py", line 347, in faster_whisper
File "faster_whisper\transcribe.py", line 689, in init
RuntimeError: CUDA failed with error CUDA-capable device(s) is/are busy or unavailable
TaskCfgVTT(is_cuda=True, uuid='617e6c9a85', cache_folder='D:/win-pyvideotrans-v3.99-428/tmp/2152/617e6c9a85', target_dir='G:/John Templeton/_video_out/John Templeton On Finding True Value 1989-mp4', source_language='英语', source_language_code='en', source_sub='G:/John Templeton/_video_out/John Templeton On Finding True Value 1989-mp4/en.srt', source_wav='D:/win-pyvideotrans-v3.99-428/tmp/2152/617e6c9a85/en.wav', source_wav_output='G:/John Templeton/_video_out/John Templeton On Finding True Value 1989-mp4/en.m4a', target_language='简体中文', target_language_code='zh-cn', target_sub='G:/John Templeton/_video_out/John Templeton On Finding True Value 1989-mp4/zh-cn.srt', target_wav='D:/win-pyvideotrans-v3.99-428/tmp/2152/617e6c9a85/target.wav', target_wav_output='G:/John Templeton/_video_out/John Templeton On Finding True Value 1989-mp4/zh-cn.m4a', name='G:/John Templeton/John Templeton On Finding True Value 1989.mp4', noextname='John Templeton On Finding True Value 1989', basename='John Templeton On Finding True Value 1989.mp4', ext='mp4', dirname='G:/John Templeton', shound_del_name=None, translate_type=13, tts_type=0, volume='+0%', pitch='+0Hz', voice_rate='+0%', voice_role='No', voice_autorate=True, video_autorate=False, remove_silent_mid=False, align_sub_audio=True, detect_language='en', recogn_type=0, model_name='large-v3', shibie_audio=None, remove_noise=False, enable_diariz=False, nums_diariz=0, rephrase=0, fix_punc=False, subtitle_language=None, app_mode='biaozhun', subtitles='', targetdir_mp4='G:/John Templeton/_video_out/John Templeton On Finding True Value 1989-mp4/John Templeton On Finding True Value 1989.mp4', novoice_mp4='D:/win-pyvideotrans-v3.99-428/tmp/2152/617e6c9a85/novoice.mp4', is_separate=False, embed_bgm=True, instrument=None, vocal='D:/win-pyvideotrans-v3.99-428/tmp/2152/617e6c9a85/vocal.wav', back_audio='', clear_cache=False, background_music=None, subtitle_type=3, only_out_mp4=True, recogn2pass=True, output_srt=2, copysrt_rawvideo=False)


=
system:Windows-10-10.0.26200-SP0
version:v3.99
frozen:True
language:zh
root_dir:D:/win-pyvideotrans-v3.99-428

1 Replies

该错误是由于 CUDA 显卡被其他程序占用、显存不足或当前 GPU 状态不可用导致的。你的配置中开启了 CUDA 加速并选用了 large-v3 模型(需 10GB+ 显存),且启用了“二次识别”(recogn2pass=True),这会进一步推高显存占用。

解决方案(按顺序尝试):

  1. 关闭“二次识别”:在主界面右上角取消勾选“二次识别”复选框。该功能会在配音后再次调用语音识别,容易与首次识别争抢显存。
  2. 确保 GPU 空闲:关闭其他可能占用显卡的大型程序(如游戏、3D 软件、AI 训练等),或重启电脑后仅运行本软件。
  3. 更换更小的模型:将语音识别的模型从 large-v3 改为 large-v3-turbomediumsmall,大幅降低显存要求。
  4. 调整高级参数的精度(点击菜单 → 工具/选项高级选项):

    • CUDA数据类型:从 float32 改为 int8(牺牲少量精度,大幅降低显存占用)。
    • beam_size:改为 1
    • best_of:改为 1
    • 启用上下文感知:设为 false
  5. 如果以上均无效,可以临时关闭 CUDA 加速,使用 CPU 模式(虽然速度慢,但能应急完成工作)。

请查阅相关文档:

Post Your Reply
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.

Related resource