#3628 TaskCfgVTT(is_cuda=True, uuid='6bf209eb1f', cache_folder='E:/localai/pyvideotrans/tmp/20292/6bf209eb1f', target_dir='MIA

2001:ee0* Posted at: 3 hours ago 👁17

ASR Error [faster-whisper (Local)] An error has occurred. There may be insufficient memory or video memory.
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
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: An error has occurred. There may be insufficient memory or video memory.
A process in the process pool was terminated abruptly while the future was running or pending.
TaskCfgVTT(is_cuda=True, uuid='6bf209eb1f', cache_folder='E:/localai/pyvideotrans/tmp/20292/6bf209eb1f', target_dir='MIAA-705-mp4', source_language='Japanese', source_language_code='ja', source_sub='MIAA-705-mp4/ja.srt', source_wav='E:/localai/pyvideotrans/tmp/20292/6bf209eb1f/ja.wav', source_wav_output='MIAA-705-mp4/ja.m4a', target_language='English', target_language_code='en', target_sub='MIAA-705-mp4/en.srt', target_wav='E:/localai/pyvideotrans/tmp/20292/6bf209eb1f/target.wav', target_wav_output='MIAA-705-mp4/en.m4a', name='E:/New folder/MIAA-705.mp4', noextname='MIAA-705', basename='MIAA-705.mp4', ext='mp4', dirname='E:/New folder', shound_del_name=None, translate_type=0, 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=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='tiqu', subtitles='', targetdir_mp4='MIAA-705-mp4/MIAA-705.mp4', novoice_mp4='E:/localai/pyvideotrans/tmp/20292/6bf209eb1f/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=False)

=
system:Windows-10-10.0.19045-SP0
version:v3.98
frozen:True
language:en
root_dir:E:/localai/pyvideotrans

1 Replies

Floor #1 DeepSeek AI reply at 3 hours ago

This error is caused by insufficient GPU memory (VRAM) when using the faster-whisper model with CUDA acceleration enabled.

The error message "A process in the process pool was terminated abruptly" and "insufficient memory or video memory" indicates that the process running the faster-whisper large-v3-turbo model on your GPU was killed because it ran out of VRAM.

Based on your configuration (is_cuda=True, model_name='large-v3-turbo'), here are the steps to resolve this:

  1. Reduce GPU Memory Usage:

    • Go to Menu -> Tools/Options -> Advanced Options.
    • In the General Settings section, set GPU同时任务数[重启生效] (GPU Concurrent Tasks) to 1. This ensures only one task uses the GPU at a time.

    • In the Speech Recognition Parameters section:

      • Set 识别准确度beam_size (Beam Size) to 1 (lowest memory usage).
      • Set 识别准确度best_of (Best Of) to 1 (lowest memory usage).
      • Set CUDA数据类型 (CUDA Data Type) to int8. This uses quantization, significantly reducing VRAM usage at a slight cost to accuracy.
    • Restart the software for the GPU task setting to take effect.

  2. Use a Smaller Model (if step 1 is insufficient):

    • On the main translation interface, in the Speech Recognition line, change the model from large-v3-turbo to large-v3, large-v2, or medium. Smaller models require less VRAM.

  3. Disable CUDA Acceleration (Last Resort):

    • If your GPU has very limited VRAM (e.g., less than 4GB), you may need to disable CUDA acceleration entirely for this task.
    • On the main translation interface, uncheck the CUDA加速 (CUDA Acceleration) checkbox. This will force the model to run on your CPU (RAM), which is slower but has more memory available. Ensure your system RAM is at least 16GB for the large-v3 model.

  4. Check Available VRAM:

    • Open Command Prompt and run nvidia-smi to see your total GPU VRAM and current usage. The large-v3-turbo model typically requires more than 8GB of free VRAM for stable operation with default settings.

Please refer to the relevant documentation:

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