#3951 语音识别阶段出错 [openai-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 anymoreTr

103.220* Posted at: 7 hours ago 👁9

语音识别阶段出错 [openai-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 31, in _exec
File "videotrans\recognition\_overall.py", line 73, in _openai
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='4d1f99db44', cache_folder='E:/translate/win-pyvideotrans-v3.98-314/tmp/2268/4d1f99db44', target_dir='F:/115/sm/nosub/MIRD-237/_video_out/r18ss.com@MIRD-237-mp4', source_language='日语', source_language_code='ja', source_sub='F:/115/sm/nosub/MIRD-237/_video_out/r18ss.com@MIRD-237-mp4/ja.srt', source_wav='E:/translate/win-pyvideotrans-v3.98-314/tmp/2268/4d1f99db44/ja.wav', source_wav_output='F:/115/sm/nosub/MIRD-237/_video_out/r18ss.com@MIRD-237-mp4/ja.m4a', target_language='简体中文', target_language_code='zh-cn', target_sub='F:/115/sm/nosub/MIRD-237/_video_out/r18ss.com@MIRD-237-mp4/zh-cn.srt', target_wav='E:/translate/win-pyvideotrans-v3.98-314/tmp/2268/4d1f99db44/target.wav', target_wav_output='F:/115/sm/nosub/MIRD-237/_video_out/r18ss.com@MIRD-237-mp4/zh-cn.m4a', name='F:/115/sm/nosub/MIRD-237/[email protected]4', noextname='r18ss.com@MIRD-237', basename='[email protected]4', ext='mp4', dirname='//RaiDrive-11524/WebDAV/115/sm/nosub/MIRD-237', shound_del_name=None, translate_type=4, tts_type=0, 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='ja', recogn_type=1, model_name='medium', 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='F:/115/sm/nosub/MIRD-237/_video_out/r18ss.com@MIRD-237-mp4/[email protected]4', novoice_mp4='E:/translate/win-pyvideotrans-v3.98-314/tmp/2268/4d1f99db44/novoice.mp4', is_separate=False, instrument=None, vocal=None, back_audio='', clear_cache=False, background_music=None, subtitle_type=0, only_out_mp4=False, recogn2pass=True, output_srt=0, copysrt_rawvideo=True)

1 Replies

这是由于显存或内存不足导致语音识别进程崩溃。

解决方案:

  1. 降低模型大小:将语音识别模型从 medium 更换为更小的 smallbase 模型。
  2. 调整高级设置:在菜单栏 工具/选项 -> 高级选项 中进行如下修改,以降低显存/内存占用:

    • 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. 检查二次识别:确认是否勾选了主界面右上角的“二次识别”。如果同时处理多个视频,二次识别会显著增加显存需求。建议先取消勾选“二次识别”后重试。
  4. 检查视频路径:你的输入文件路径 F:/115/sm/nosub/MIRD-237/[email protected] 较长且包含特殊符号 @,这可能导致处理异常。请将视频文件重命名为简短英文名(如 test.mp4),并移动到更浅的目录(如 D:\video)再尝试处理。
  5. 串行处理:如果你在批量处理多个视频,请在 工具/选项 -> 高级选项 中勾选 批量翻译时强制串行,避免多个任务同时竞争资源。

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