#3913 TaskCfgSTT(is_cuda=True, uuid='2a93e3f3d2', cache_folder='F:/迅雷下载/win-pyvideotrans-v3.98-329/tmp/12240/2a93e3f3d2', targ

123.145* Posted at: 4 hours ago 👁6

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-turbo 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\_speech2text.py", line 159, 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:large-v3-turbo 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
TaskCfgSTT(is_cuda=True, uuid='2a93e3f3d2', cache_folder='F:/迅雷下载/win-pyvideotrans-v3.98-329/tmp/12240/2a93e3f3d2', target_dir='F:/迅雷下载/win-pyvideotrans-v3.98-329/output/recogn', source_language=None, source_language_code=None, source_sub=None, source_wav=None, source_wav_output=None, target_language=None, target_language_code=None, target_sub='F:/迅雷下载/win-pyvideotrans-v3.98-329/output/recogn/5.srt', target_wav=None, target_wav_output=None, name='F:/兼职导出/字幕音频/4.2-recdot-马来语/5.wav', noextname='5', basename='5.wav', ext='wav', dirname='F:/兼职导出/字幕音频/4.2-recdot-马来语', shound_del_name=None, detect_language='ms', recogn_type=0, model_name='large-v3-turbo', shibie_audio='F:/迅雷下载/win-pyvideotrans-v3.98-329/tmp/12240/2a93e3f3d2/5-1775139375.476247.wav', remove_noise=False, enable_diariz=False, nums_diariz=0, rephrase=0, fix_punc=False)


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

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-turbo 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. 更换更小的模型:将语音识别模型从 large-v3-turbo 更换为更小的模型,例如 mediumsmallbaselarge-v3-turbo 模型对显存和内存要求较高。
  2. 调整高级设置以降低资源占用:在软件顶部菜单栏,进入 工具/选项 -> 高级选项,在“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 (未选中)。
  3. 检查并限制并发任务:在“高级选项”的“通用设置”区域,确保 GPU同时任务数 设置为 1(除非你的单张显卡显存大于20GB)。
  4. 关闭其他占用GPU的程序:在进行语音识别前,关闭所有可能占用大量显存的应用程序(如游戏、其他AI软件、浏览器等)。
  5. 使用纯CPU模式:如果显卡显存确实太小(例如小于4GB),可以尝试取消勾选主界面右上角的 CUDA加速 选项,完全依赖CPU进行识别(速度会慢很多)。

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