#4071 TaskCfgVTT(is_cuda=True, uuid='2bd78abc16', cache_folder='D:/win-pyvideotrans-v3.98-329/tmp/28944/2bd78abc16', target_di

91.199* Posted at: 2 hours ago 👁13

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:base GPU0
Traceback (most recent call last):
File "videotrans\configure\_base.py", line 281, in _new_process
File "videotrans\process\signelobj.py", line 80, 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

Traceback (most recent call last):
File "videotrans\configure\_base.py", line 281, in _new_process
File "videotrans\process\signelobj.py", line 80, 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 360, in recogn
File "videotrans\recognition\__init__.py", line 272, 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:base GPU0
Traceback (most recent call last):
File "videotrans\configure\_base.py", line 281, in _new_process
File "videotrans\process\signelobj.py", line 80, 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
TaskCfgVTT(is_cuda=True, uuid='2bd78abc16', cache_folder='D:/win-pyvideotrans-v3.98-329/tmp/28944/2bd78abc16', target_dir='/ymnas/跨境/tk视频/丰唇膏/已剪/无字幕/西班牙语/0414-1-1-mp4', source_language='英语', source_language_code='en', source_sub='/ymnas/跨境/tk视频/丰唇膏/已剪/无字幕/西班牙语/0414-1-1-mp4/en.srt', source_wav='D:/win-pyvideotrans-v3.98-329/tmp/28944/2bd78abc16/en.wav', source_wav_output='/ymnas/跨境/tk视频/丰唇膏/已剪/无字幕/西班牙语/0414-1-1-mp4/en.m4a', target_language='西班牙语', target_language_code='es', target_sub='/ymnas/跨境/tk视频/丰唇膏/已剪/无字幕/西班牙语/0414-1-1-mp4/es.srt', target_wav='D:/win-pyvideotrans-v3.98-329/tmp/28944/2bd78abc16/target.wav', target_wav_output='/ymnas/跨境/tk视频/丰唇膏/已剪/无字幕/西班牙语/0414-1-1-mp4/es.m4a', name='Z:/tk视频/丰唇膏/已剪/无字幕/英文/0414-1-1.mp4', noextname='0414-1-1', basename='0414-1-1.mp4', ext='mp4', dirname='//ymnas/跨境/tk视频/丰唇膏/已剪/无字幕/英文', shound_del_name=None, translate_type=0, tts_type=0, volume='+0%', pitch='+0Hz', voice_rate='+0%', voice_role='Alonso(Male/US)', voice_autorate=True, video_autorate=True, remove_silent_mid=False, align_sub_audio=True, detect_language='en', recogn_type=0, model_name='base', 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='/ymnas/跨境/tk视频/丰唇膏/已剪/无字幕/西班牙语/0414-1-1-mp4/0414-1-1.mp4', novoice_mp4='D:/win-pyvideotrans-v3.98-329/tmp/28944/2bd78abc16/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=False, output_srt=0, copysrt_rawvideo=False)


=
system:Windows-10-10.0.19045-SP0
version:v3.98
frozen:True
language:zh
root_dir:D:/win-pyvideotrans-v3.98-329

1 Replies

这是由于显存或内存不足导致的语音识别进程崩溃。

解决方案:

  1. 检查并调整 GPU 设置

    • 确认已正确安装 CUDA 12.8 和 cuDNN 9.x。
    • 在软件主界面右上角,取消勾选“二次识别”。同时处理多个视频时,二次识别会占用大量显存。
    • 菜单 -> 工具/选项 -> 高级选项 中,将 GPU同时任务数[重启生效] 设为 1,然后重启软件。
  2. 降低模型和参数以节省资源

    • 将语音识别模型从 base 更换为更小的 tiny 模型。
    • 高级选项 -> 语音识别参数 中,进行以下调整:

      • CUDA数据类型:从 float32 改为 float16int8
      • 识别准确度beam_size:从 5 改为 1
      • 识别准确度best_of:从 5 改为 1
      • 启用上下文感知:确保为 false
  3. 检查文件路径

    • 你的输入文件路径 Z:/tk视频/... 和输出路径 /ymnas/跨境/... 包含中文且层级较深,这可能导致 Windows 路径长度超限或权限问题。建议将视频文件复制到本地简单目录(如 D:\video)并重命名为英文短名后再处理。
  4. 升级软件版本

    • 你使用的是 v3.98 版本,可以尝试升级到最新的 v3.99 版本,该版本在多显卡环境下能自动选择可用显存最大的显卡。

如果上述步骤仍无法解决问题,可能是视频过长或硬件资源确实不足,请尝试使用纯 CPU 模式(取消勾选主界面的“CUDA加速”),但处理速度会显著下降。

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