#3614 TaskCfgVTT(is_cuda=True, uuid='83ed049529', cache_folder='G:/VideoTrans/tmp/5720/83ed049529', target_dir='D:/Bl tuatriol

113.88* Posted at: 1 day ago 👁18

合并阶段出错 出错了,可能内存或显存不足
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 288, in run
File "videotrans\task\trans_create.py", line 789, in assembling
File "videotrans\task\trans_create.py", line 1251, in _join_video_audio_srt
File "videotrans\task\trans_create.py", line 530, in recogn2pass
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: 出错了,可能内存或显存不足
A process in the process pool was terminated abruptly while the future was running or pending.
TaskCfgVTT(is_cuda=True, uuid='83ed049529', cache_folder='G:/VideoTrans/tmp/5720/83ed049529', target_dir='D:/Bl tuatriol/Factory 2025/2025/6 - Modeling the Second Module-mp4', source_language='英语', source_language_code='en', source_sub='D:/Bl tuatriol/Factory 2025/2025/6 - Modeling the Second Module-mp4/en.srt', source_wav='G:/VideoTrans/tmp/5720/83ed049529/en.wav', source_wav_output='D:/Bl tuatriol/Factory 2025/2025/6 - Modeling the Second Module-mp4/en.m4a', target_language='简体中文', target_language_code='zh-cn', target_sub='D:/Bl tuatriol/Factory 2025/2025/6 - Modeling the Second Module-mp4/zh-cn.srt', target_wav='G:/VideoTrans/tmp/5720/83ed049529/target.wav', target_wav_output='D:/Bl tuatriol/Factory 2025/2025/6 - Modeling the Second Module-mp4/zh-cn.m4a', name='D:/Bl tuatriol/Factory 2025/6 - Modeling the Second Module.mp4', noextname='6 - Modeling the Second Module', basename='6 - Modeling the Second Module.mp4', ext='mp4', dirname='D:/Bl tuatriol/Factory 2025', shound_del_name=None, translate_type=0, tts_type=10, volume='+0%', pitch='+0Hz', voice_rate='+0%', voice_role='001.wav', voice_autorate=False, video_autorate=False, remove_silent_mid=False, align_sub_audio=True, detect_language='en', 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='D:/Bl tuatriol/Factory 2025/2025/6 - Modeling the Second Module-mp4/6 - Modeling the Second Module.mp4', novoice_mp4='G:/VideoTrans/tmp/5720/83ed049529/novoice.mp4', is_separate=False, instrument=None, vocal=None, back_audio='', clear_cache=False, background_music=None, subtitle_type=1, 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:zh
root_dir:G:/VideoTrans

1 Replies

该错误是由于系统内存或显存不足,导致进程池中的进程在执行二次识别(recogn2pass)时被意外终止。

根据日志,问题发生在视频翻译的合并阶段,具体是在进行二次识别(recogn2pass=True)时,调用了 faster-whisper 模型(model_name='large-v3-turbo')并启用了 CUDA 加速(is_cuda=True),进程因资源不足而崩溃。

解决方案如下:

  1. 降低资源消耗

    • 在视频翻译主界面,取消勾选 二次识别 选项。这是最直接的解决方法,可以避免在合成阶段再次进行高负载的语音识别。
    • 如果仍需二次识别,请进入 菜单 -> 工具 -> 高级选项,在 【语音识别参数】 区域,将 识别准确度beam_size识别准确度best_of 都设置为 1,以降低显存和内存占用。
    • 【通用设置】 区域,确保 GPU同时任务数[重启生效] 设置为 1
  2. 调整识别模型

    • 如果必须进行二次识别,考虑在首次语音识别时使用更小的模型(例如 smallmedium),或者在二次识别时手动切换到更小的模型(但这需要修改任务配置,较为复杂)。最简便的方法是直接关闭二次识别。
  3. 检查系统资源

    • 在处理视频时,打开任务管理器,监控内存和GPU显存的使用情况。如果接近满载,请关闭其他占用资源的程序。
    • 确保虚拟内存页面文件大小设置充足。
  4. 优化高级设置

    • 进入 菜单 -> 工具 -> 高级选项,在 【通用设置】 区域,将 最大进程数[重启生效] 适当调低(例如设置为CPU核心数的一半)。
    • 【字幕配音调整】 区域,将 并发配音线程数 设为 1,减少并行任务对资源的争抢。

由于错误明确指向资源不足,优先采用方案1(关闭二次识别)通常能立即解决问题。如果关闭后仍出现类似错误,则需按照方案4全面降低软件并发数。

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