#4092 TaskCfgVTT(is_cuda=True, uuid='5ccdb57e4b', cache_folder='F:/win-pyvideotrans-v3.99/tmp/28884/5ccdb57e4b', target_dir='D

240e:b8f* Posted at: 2 days ago 👁26

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:large-v3 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:large-v3 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='5ccdb57e4b', cache_folder='F:/win-pyvideotrans-v3.99/tmp/28884/5ccdb57e4b', target_dir='D:/Season 02/_video_out/S2 (3)-mkv', source_language='英语', source_language_code='en', source_sub='D:/Season 02/_video_out/S2 (3)-mkv/en.srt', source_wav='F:/win-pyvideotrans-v3.99/tmp/28884/5ccdb57e4b/en.wav', source_wav_output='D:/Season 02/_video_out/S2 (3)-mkv/en.m4a', target_language='简体中文', target_language_code='zh-cn', target_sub='D:/Season 02/_video_out/S2 (3)-mkv/zh-cn.srt', target_wav='F:/win-pyvideotrans-v3.99/tmp/28884/5ccdb57e4b/target.wav', target_wav_output='D:/Season 02/_video_out/S2 (3)-mkv/zh-cn.m4a', name='D:/Season 02/S2 (3).mkv', noextname='S2 (3)', basename='S2 (3).mkv', ext='mkv', dirname='D:/Season 02', shound_del_name=None, translate_type=13, tts_type=0, volume='+0%', pitch='+0Hz', voice_rate='+0%', voice_role='No', voice_autorate=True, video_autorate=False, remove_silent_mid=False, align_sub_audio=True, detect_language='en', recogn_type=0, model_name='large-v3', 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='D:/Season 02/_video_out/S2 (3)-mkv/S2 (3).mp4', novoice_mp4='F:/win-pyvideotrans-v3.99/tmp/28884/5ccdb57e4b/novoice.mp4', is_separate=True, embed_bgm=True, instrument='F:/win-pyvideotrans-v3.99/tmp/28884/5ccdb57e4b/instrument.wav', vocal='F:/win-pyvideotrans-v3.99/tmp/28884/5ccdb57e4b/vocal.wav', back_audio='', clear_cache=True, background_music=None, subtitle_type=3, only_out_mp4=True, recogn2pass=True, output_srt=2, copysrt_rawvideo=False)


=
system:Windows-10-10.0.26200-SP0
version:v3.99
frozen:True
language:zh
root_dir:F:/win-pyvideotrans-v3.99

1 Replies

这是由于显存或内存不足导致的。错误信息明确指出在使用 faster-whisper(本地)large-v3 模型时,GPU0 的显存或系统内存不足以完成语音识别任务。

解决方案:

  1. 降低模型大小:在语音识别渠道的下拉菜单中,将模型从 large-v3 更换为更小的模型,例如 mediumsmallbaselarge-v3 模型至少需要 10GB 显存。
  2. 调整高级设置以降低显存占用:点击 菜单 -> 工具/选项 -> 高级选项,在“语音识别参数”区域进行以下调整:

    • 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. 检查并关闭二次识别:从日志中看到 recogn2pass=True,表示开启了二次识别。如果同时处理多个视频,二次识别会占用额外显存。请在主界面右上角取消勾选 二次识别 选项。
  4. 检查显卡选择:如果你有多块显卡,且第一块显卡(GPU0)的可用显存较小,软件可能因默认使用它而报错。v3.99 版本已优化为默认使用可用显存最大的显卡。请确保你的软件版本为 v3.99。
  5. 确保CUDA环境正确:确认已安装 CUDA 12.8 和 cuDNN 9.x 及以上版本,并在主界面勾选了 CUDA加速

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