#3774 TaskCfgVTT(is_cuda=True, uuid='9e8118ff0e', cache_folder='E:/pyvideotrans-v3.98/tmp/25924/9e8118ff0e', target_dir='I:/短片

2403:18c0* 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 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 284, in _new_process
File "concurrent\futures\_base.py", line 458, in result
File "concurrent\futures\_base.py", line 403, in __get_result
File "videotrans\configure\_base.py", line 284, 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 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 process in the process pool was terminated abruptly while the future was running or pending.
TaskCfgVTT(is_cuda=True, uuid='9e8118ff0e', cache_folder='E:/pyvideotrans-v3.98/tmp/25924/9e8118ff0e', target_dir='I:/短片/_video_out/9.二战海军水雷是如何工作的-mp4', source_language='英语', source_language_code='en', source_sub='I:/短片/_video_out/9.二战海军水雷是如何工作的-mp4/en.srt', source_wav='E:/pyvideotrans-v3.98/tmp/25924/9e8118ff0e/en.wav', source_wav_output='I:/短片/_video_out/9.二战海军水雷是如何工作的-mp4/en.m4a', target_language='简体中文', target_language_code='zh-cn', target_sub='I:/短片/_video_out/9.二战海军水雷是如何工作的-mp4/zh-cn.srt', target_wav='E:/pyvideotrans-v3.98/tmp/25924/9e8118ff0e/target.wav', target_wav_output='I:/短片/_video_out/9.二战海军水雷是如何工作的-mp4/zh-cn.m4a', name='I:/短片/处理中1-10/9.二战海军水雷是如何工作的.mp4', noextname='9.二战海军水雷是如何工作的', basename='9.二战海军水雷是如何工作的.mp4', ext='mp4', dirname='I:/短片/处理中1-10', shound_del_name=None, translate_type=10, tts_type=0, volume='+0%', pitch='+0Hz', voice_rate='+0%', voice_role='Yunjian(Male/CN)', voice_autorate=False, video_autorate=True, 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='I:/短片/_video_out/9.二战海军水雷是如何工作的-mp4/9.二战海军水雷是如何工作的.mp4', novoice_mp4='E:/pyvideotrans-v3.98/tmp/25924/9e8118ff0e/novoice.mp4', is_separate=False, instrument=None, vocal=None, back_audio='', clear_cache=False, background_music=None, subtitle_type=1, only_out_mp4=False, recogn2pass=True, output_srt=0, copysrt_rawvideo=False)


=
system:Windows-10-10.0.22621-SP0
version:v3.98
frozen:True
language:zh
root_dir:E:/pyvideotrans-v3.98

1 Replies

这是由于显存或内存不足导致语音识别进程被强制终止。

错误信息明确指出“可能内存或显存不足”,并且进程池中的进程被意外终止。从配置中可以看到,您启用了CUDA加速(is_cuda=True),并使用了较大的模型 large-v3-turbo

解决方案:

  1. 降低模型大小:这是最直接有效的方法。将语音识别模型从 large-v3-turbo 更换为更小的模型,例如 mediumsmallbaselarge-v3-turbo 模型需要大量显存/内存。
  2. 调整高级设置以降低资源消耗:在软件顶部菜单栏,进入 工具/选项 -> 高级选项,找到“语音识别参数”区域,进行以下修改:

    • 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:确保此项为 false(未勾选)。
  3. 检查文件路径:您的输入文件路径 I:/短片/处理中1-10/9.二战海军水雷是如何工作的.mp4 包含中文字符。虽然当前错误是资源不足,但路径问题也可能引发其他错误。建议将视频文件重命名为纯英文或数字名称(如 9.mp4),并移动到不含中文的目录(如 D:/videos)再尝试处理。
  4. 关闭其他占用GPU/内存的程序:在处理视频时,关闭游戏、浏览器、大型设计软件等,为pyVideoTrans释放更多资源。
  5. 验证CUDA环境:虽然已启用CUDA,但仍需确认安装正确。在CMD中输入 nvidia-smi 查看显卡状态和CUDA版本。软件内置支持CUDA 12.8。

如果以上调整后问题依旧,请提供 logs 文件夹中最新日志文件的底部更多行信息,以便进一步分析。

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