#3797 TaskCfgVTT(is_cuda=True, uuid='7e0c97249c', cache_folder='D:/project/Videotrans/win-pyvideotrans-v3.97-0304/tmp/40824/7e

57.181* Posted at: 3 hours ago 👁7

语音识别阶段出错 [faster-whisper(本地)] 出错了,可能内存或显存不足
A child process terminated abruptly, the process pool is not usable anymore
Traceback (most recent call last):
File "videotrans\configure\_base.py", line 280, in _new_process
File "videotrans\process\signelobj.py", line 81, in submit_task_gpu
File "concurrent\futures\process.py", line 720, in submit
concurrent.futures.process.BrokenProcessPool: A child process terminated abruptly, the process pool is not usable anymore

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "videotrans\task\job.py", line 105, in run
File "videotrans\task\trans_create.py", line 353, in recogn
File "videotrans\recognition\__init__.py", line 265, 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 294, in _new_process
RuntimeError: 出错了,可能内存或显存不足
A child process terminated abruptly, the process pool is not usable anymore
TaskCfgVTT(is_cuda=True, uuid='7e0c97249c', cache_folder='D:/project/Videotrans/win-pyvideotrans-v3.97-0304/tmp/40824/7e0c97249c', target_dir='E:/编程/Vibe Coding/Udemy - Claude Code Masterclass Plan, Build, Test & Deploy Real App/_video_out/1. Introduction-mp4', source_language='英语', source_language_code='en', source_sub='E:/编程/Vibe Coding/Udemy - Claude Code Masterclass Plan, Build, Test & Deploy Real App/_video_out/1. Introduction-mp4/en.srt', source_wav='D:/project/Videotrans/win-pyvideotrans-v3.97-0304/tmp/40824/7e0c97249c/en.wav', source_wav_output='E:/编程/Vibe Coding/Udemy - Claude Code Masterclass Plan, Build, Test & Deploy Real App/_video_out/1. Introduction-mp4/en.m4a', target_language='简体中文', target_language_code='zh-cn', target_sub='E:/编程/Vibe Coding/Udemy - Claude Code Masterclass Plan, Build, Test & Deploy Real App/_video_out/1. Introduction-mp4/zh-cn.srt', target_wav='D:/project/Videotrans/win-pyvideotrans-v3.97-0304/tmp/40824/7e0c97249c/target.wav', target_wav_output='E:/编程/Vibe Coding/Udemy - Claude Code Masterclass Plan, Build, Test & Deploy Real App/_video_out/1. Introduction-mp4/zh-cn.m4a', name='E:/编程/Vibe Coding/Udemy - Claude Code Masterclass Plan, Build, Test & Deploy Real App/1. Introduction.mp4', noextname='1. Introduction', basename='1. Introduction.mp4', ext='mp4', dirname='E:/编程/Vibe Coding/Udemy - Claude Code Masterclass Plan, Build, Test & Deploy Real App', shound_del_name=None, translate_type=5, tts_type=10, volume='+0%', pitch='+0Hz', voice_rate='+0%', voice_role='aozi.wav', voice_autorate=True, video_autorate=False, remove_silent_mid=False, align_sub_audio=True, detect_language='en', recogn_type=0, model_name='large-v3-turbo', 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='E:/编程/Vibe Coding/Udemy - Claude Code Masterclass Plan, Build, Test & Deploy Real App/_video_out/1. Introduction-mp4/1. Introduction.mp4', novoice_mp4='D:/project/Videotrans/win-pyvideotrans-v3.97-0304/tmp/40824/7e0c97249c/novoice.mp4', is_separate=False, instrument=None, vocal=None, back_audio='', clear_cache=False, background_music=None, subtitle_type=4, only_out_mp4=False, recogn2pass=True, output_srt=2, copysrt_rawvideo=False)


=
system:Windows-10-10.0.26200-SP0
version:v3.97
frozen:True
language:zh
root_dir:D:/project/Videotrans/win-pyvideotrans-v3.97-0304

1 Replies

这是由于显存或内存不足导致语音识别进程崩溃。

根据错误信息,您在使用 faster-whisper(本地)large-v3-turbo 模型进行语音识别时,CUDA 加速已开启(is_cuda=True),但子进程异常终止,提示“可能内存或显存不足”。

解决方案:

  1. 降低模型大小:将语音识别模型从 large-v3-turbo 更换为更小的模型,例如 mediumsmallbaselarge-v3 系列模型最低需要约 10GB 显存。在软件主界面的“语音识别”行右侧下拉菜单中直接更换。
  2. 调整高级设置:在菜单栏 工具/选项 -> 高级选项 中,找到“语音识别参数”区域,进行如下修改以降低显存占用:

    • CUDA数据类型: 将 float32 改为 float16int8
    • 识别准确度beam_size: 将 5 改为 1
    • 识别准确度best_of: 将 5 改为 1
    • 启用上下文感知: 将 true 改为 false
  3. 检查任务并发:在“高级选项”的“通用设置”区域,确保 GPU同时任务数[重启生效] 设置为 1(除非您的单卡显存大于20GB)。
  4. 关闭二次识别:您当前配置中 recogn2pass=True(启用了二次识别)。在处理多个视频时,二次识别可能与后续视频的首次转录同时运行,导致需要2-3倍的显存。建议在主界面取消勾选“二次识别”选项。
  5. 简化文件路径:您的视频文件路径 E:/编程/Vibe Coding/Udemy - Claude Code Masterclass Plan, Build, Test & Deploy Real App/1. Introduction.mp4 包含中文且较长。虽然此错误主要与资源有关,但过长的路径也可能引发其他问题。建议将视频文件移动到更浅的英文目录(如 D:/videos)并重命名为简短名称(如 intro.mp4)再进行处理。

如果以上步骤仍无法解决,请考虑升级软件到最新版本(v3.98),该版本在多显卡环境下能自动选择可用显存最大的显卡。

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