#3601 TaskCfgVTT(is_cuda=True, uuid='7705dc93ac', cache_folder='E:/win-pyvideotrans-v3.98-312/tmp/15312/7705dc93ac', target_di

1.55* Posted at: 1 day ago 👁26

ASR Error [faster-whisper (Local)] An error has occurred. There may be insufficient memory or video memory.
A child process terminated abruptly, the process pool is not usable anymore
Traceback (most recent call last):
File "videotrans\configure\_base.py", line 280, in _new_process
File "videotrans\process\signelobj.py", line 81, 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 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 child process terminated abruptly, the process pool is not usable anymore
TaskCfgVTT(is_cuda=True, uuid='7705dc93ac', cache_folder='E:/win-pyvideotrans-v3.98-312/tmp/15312/7705dc93ac', target_dir='C:/Users/DreamStore/Downloads/douyin/_video_out/@奥斯28K 俄罗斯警察的搞笑日常，讽刺拉满！ 1080P-mp4', source_language='Simplified Chinese', source_language_code='zh-cn', source_sub='C:/Users/DreamStore/Downloads/douyin/_video_out/@奥斯28K 俄罗斯警察的搞笑日常，讽刺拉满！ 1080P-mp4/zh-cn.srt', source_wav='E:/win-pyvideotrans-v3.98-312/tmp/15312/7705dc93ac/zh-cn.wav', source_wav_output='C:/Users/DreamStore/Downloads/douyin/_video_out/@奥斯28K 俄罗斯警察的搞笑日常，讽刺拉满！ 1080P-mp4/zh-cn.m4a', target_language='Korean', target_language_code='ko', target_sub='C:/Users/DreamStore/Downloads/douyin/_video_out/@奥斯28K 俄罗斯警察的搞笑日常，讽刺拉满！ 1080P-mp4/ko.srt', target_wav='E:/win-pyvideotrans-v3.98-312/tmp/15312/7705dc93ac/target.wav', target_wav_output='C:/Users/DreamStore/Downloads/douyin/_video_out/@奥斯28K 俄罗斯警察的搞笑日常，讽刺拉满！ 1080P-mp4/ko.m4a', name='C:/Users/DreamStore/Downloads/douyin/@奥斯28K 俄罗斯警察的搞笑日常，讽刺拉满！ 1080P.mp4', noextname='@奥斯28K 俄罗斯警察的搞笑日常，讽刺拉满！ 1080P', basename='@奥斯28K 俄罗斯警察的搞笑日常，讽刺拉满！ 1080P.mp4', ext='mp4', dirname='C:/Users/DreamStore/Downloads/douyin', shound_del_name=None, translate_type=3, tts_type=0, volume='+0%', pitch='+0Hz', voice_rate='+0%', voice_role='HyunsuMultilingual(Male/KR)', voice_autorate=True, video_autorate=False, remove_silent_mid=False, align_sub_audio=True, detect_language='zh-cn', 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='C:/Users/DreamStore/Downloads/douyin/_video_out/@奥斯28K 俄罗斯警察的搞笑日常，讽刺拉满！ 1080P-mp4/@奥斯28K 俄罗斯警察的搞笑日常，讽刺拉满！ 1080P.mp4', novoice_mp4='E:/win-pyvideotrans-v3.98-312/tmp/15312/7705dc93ac/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:/win-pyvideotrans-v3.98-312

1 Replies

Floor #1 DeepSeek AI reply at 1 day ago

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

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

To resolve this, follow these steps:

  1. Reduce GPU Memory Usage:

    • In the main interface, change the faster-whisper model from large-v3-turbo to a smaller one, such as large-v3, large-v2, or medium. The large-v3-turbo model requires significant VRAM.
    • Go to Menu -> Tools/Options -> Advanced Options -> Speech Recognition Parameters.
    • Reduce the beam_size and best_of values to 1 (lowest memory consumption, lower accuracy).
    • Change the CUDA Data Type to int8 (consumes fewer resources, faster, lower precision).

  2. Limit Concurrent GPU Tasks:

    • In Menu -> Tools/Options -> Advanced Options -> General Settings, ensure GPU Simultaneous Tasks [Restart Effective] is set to 1. This prevents multiple GPU tasks from competing for memory.

  3. Check for Path/File Name Issues:

    • Your input file path is extremely long and contains special characters (@, !, Chinese characters). This can cause command-line length issues in Windows.
    • Rename the input video file to a short, simple name using only English letters and numbers (e.g., video.mp4). Move it to a shallow directory like C:\test\.
    • Ensure your software installation path (E:/win-pyvideotrans-v3.98-312/) is not too deep.

  4. If the above fails, disable CUDA acceleration (as a last resort):

    • Uncheck the CUDA Acceleration option in the main interface. This will force the model to run on the CPU, which is much slower but uses system RAM instead of VRAM. Ensure you have at least 16GB of system RAM for the large-v3 model.

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