#4090 TaskCfgVTT(is_cuda=True, uuid='b1eb6c95f8', cache_folder='D:/pyvideo/pyvideotrans/tmp/34740/b1eb6c95f8', target_dir='D:/

182.100* Posted at: 2 days ago 👁25

语音识别阶段出错 [faster-whisper(本地)] 出错了,可能内存或显存不足 Model:small.en GPU0
Traceback (most recent call last):
File "videotrans\configure\_base.py", line 285, in _new_process
File "concurrent\futures\_base.py", line 458, in result
File "concurrent\futures\_base.py", line 403, in __get_result
concurrent.futures.process.BrokenProcessPool: A process in the process pool was terminated abruptly while the future was running or pending.

Traceback (most recent call last):
File "videotrans\configure\_base.py", line 285, in _new_process
File "concurrent\futures\_base.py", line 458, in result
File "concurrent\futures\_base.py", line 403, in __get_result
concurrent.futures.process.BrokenProcessPool: A process in the process pool was terminated abruptly while the future was running or pending.

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 360, in recogn
File "videotrans\recognition\__init__.py", line 292, 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 303, in _new_process
RuntimeError: 出错了,可能内存或显存不足 Model:small.en GPU0
Traceback (most recent call last):
File "videotrans\configure\_base.py", line 285, in _new_process
File "concurrent\futures\_base.py", line 458, in result
File "concurrent\futures\_base.py", line 403, in __get_result
concurrent.futures.process.BrokenProcessPool: A process in the process pool was terminated abruptly while the future was running or pending.
TaskCfgVTT(is_cuda=True, uuid='b1eb6c95f8', cache_folder='D:/pyvideo/pyvideotrans/tmp/34740/b1eb6c95f8', target_dir='D:/test/_video_out/11. Closing Thoughts-mp4', source_language='英语', source_language_code='en', source_sub='D:/test/_video_out/11. Closing Thoughts-mp4/en.srt', source_wav='D:/pyvideo/pyvideotrans/tmp/34740/b1eb6c95f8/en.wav', source_wav_output='D:/test/_video_out/11. Closing Thoughts-mp4/en.m4a', target_language='简体中文', target_language_code='zh-cn', target_sub='D:/test/_video_out/11. Closing Thoughts-mp4/zh-cn.srt', target_wav='D:/pyvideo/pyvideotrans/tmp/34740/b1eb6c95f8/target.wav', target_wav_output='D:/test/_video_out/11. Closing Thoughts-mp4/zh-cn.m4a', name='D:/test/11. Closing Thoughts.mp4', noextname='11. Closing Thoughts', basename='11. Closing Thoughts.mp4', ext='mp4', dirname='D:/test', shound_del_name=None, translate_type=4, tts_type=1, volume='+0%', pitch='+0Hz', voice_rate='+0%', voice_role='No', voice_autorate=False, video_autorate=False, remove_silent_mid=False, align_sub_audio=True, detect_language='en', recogn_type=0, model_name='small.en', shibie_audio=None, remove_noise=False, enable_diariz=False, nums_diariz=0, rephrase=0, fix_punc=False, subtitle_language=None, app_mode='tiqu', subtitles='', targetdir_mp4='D:/test/_video_out/11. Closing Thoughts-mp4/11. Closing Thoughts.mp4', novoice_mp4='D:/pyvideo/pyvideotrans/tmp/34740/b1eb6c95f8/novoice.mp4', is_separate=True, embed_bgm=True, instrument='D:/pyvideo/pyvideotrans/tmp/34740/b1eb6c95f8/instrument.wav', vocal='D:/pyvideo/pyvideotrans/tmp/34740/b1eb6c95f8/vocal.wav', back_audio='', clear_cache=True, background_music=None, subtitle_type=0, only_out_mp4=True, recogn2pass=True, output_srt=2, copysrt_rawvideo=False)


=
system:Windows-10-10.0.19045-SP0
version:v3.99
frozen:True
language:zh
root_dir:D:/pyvideo/pyvideotrans

1 Replies

这是由于内存或显存不足导致的进程池异常终止。

  1. 检查是否启用了“二次识别”。从日志 recogn2pass=True 可见已开启。请首先在主界面右上角取消勾选“二次识别”。同时处理多个视频时,二次识别会与首次转录并行,导致显存需求倍增。
  2. 如果问题依旧,请降低语音识别模型大小。当前使用的是 small.en 模型,可尝试更换为更小的 base.entiny.en 模型。
  3. 调整高级设置以降低资源占用。进入 菜单 -> 工具/选项 -> 高级选项

    • 在“语音识别参数”区域,将 CUDA数据类型float32 改为 float16int8
    • 识别准确度beam_size识别准确度best_of5 改为 1
    • 启用上下文感知 设为 false
  4. 确保 CUDA 环境正确。在 CMD 中输入 nvcc -Vnvidia-smi 确认 CUDA 版本为 12.8 且显卡驱动正常。
  5. 如果以上步骤无效,可能是视频过长或系统内存不足。尝试关闭其他占用内存的软件,或使用更小的模型(如 tiny.en)。

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