#1776 TaskCfg(cache_folder='I:/fanyi/win-pyvideotrans-v3.87/tmp/34840/27d82477d4', target_dir='I:/账号运营工作/BaiduSyncdisk/视频待投递/G

1.162* Posted at: 4 days ago 👁25

语音识别阶段出错:[faster-whisper(本地)] 运行时错误:Traceback (most recent call last):
File "videotrans\process\_overall.py", line 149, in run
File "faster_whisper\transcribe.py", line 1851, in restore_speech_timestamps
File "faster_whisper\transcribe.py", line 1190, in generate_segments
File "faster_whisper\transcribe.py", line 1400, in encode
RuntimeError: CUDA failed with error out of memory
:
Traceback (most recent call last):
File "videotrans\task\job.py", line 113, in run
File "videotrans\task\trans_create.py", line 360, in recogn
File "videotrans\recognition\__init__.py", line 243, in run
File "videotrans\recognition\_base.py", line 78, in run
File "videotrans\recognition\_overall.py", line 180, in _exec
RuntimeError: Traceback (most recent call last):
File "videotrans\process\_overall.py", line 149, in run
File "faster_whisper\transcribe.py", line 1851, in restore_speech_timestamps
File "faster_whisper\transcribe.py", line 1190, in generate_segments
File "faster_whisper\transcribe.py", line 1400, in encode
RuntimeError: CUDA failed with error out of memory

TaskCfg(cache_folder='I:/fanyi/win-pyvideotrans-v3.87/tmp/34840/27d82477d4', target_dir='I:/账号运营工作/BaiduSyncdisk/视频待投递/Granate旅行生活/_video_out/Granate Travelling walking indoor, #Hallstatt hotel, shorts, thigh high boots, high heels-mp4', remove_noise=False, is_separate=False, detect_language='en', subtitle_language=None, source_language='英语', target_language='简体中文', source_language_code='en', target_language_code='zh-cn', source_sub='I:/账号运营工作/BaiduSyncdisk/视频待投递/Granate旅行生活/_video_out/Granate Travelling walking indoor, #Hallstatt hotel, shorts, thigh high boots, high heels-mp4/en.srt', target_sub='I:/账号运营工作/BaiduSyncdisk/视频待投递/Granate旅行生活/_video_out/Granate Travelling walking indoor, #Hallstatt hotel, shorts, thigh high boots, high heels-mp4/zh-cn.srt', source_wav='I:/fanyi/win-pyvideotrans-v3.87/tmp/34840/27d82477d4/en.wav', source_wav_output='I:/账号运营工作/BaiduSyncdisk/视频待投递/Granate旅行生活/_video_out/Granate Travelling walking indoor, #Hallstatt hotel, shorts, thigh high boots, high heels-mp4/en.m4a', target_wav='I:/fanyi/win-pyvideotrans-v3.87/tmp/34840/27d82477d4/target.wav', target_wav_output='I:/账号运营工作/BaiduSyncdisk/视频待投递/Granate旅行生活/_video_out/Granate Travelling walking indoor, #Hallstatt hotel, shorts, thigh high boots, high heels-mp4/zh-cn.m4a', subtitles='', novoice_mp4='I:/fanyi/win-pyvideotrans-v3.87/tmp/34840/27d82477d4/novoice.mp4', noextname='Granate Travelling walking indoor, #Hallstatt hotel, shorts, thigh high boots, high heels', shibie_audio=None, targetdir_mp4='I:/账号运营工作/BaiduSyncdisk/视频待投递/Granate旅行生活/_video_out/Granate Travelling walking indoor, #Hallstatt hotel, shorts, thigh high boots, high heels-mp4/Granate Travelling walking indoor, #Hallstatt hotel, shorts, thigh high boots, high heels.mp4', instrument=None, vocal=None, back_audio='', background_music=None, app_mode='biaozhun', subtitle_type=1, volume='+0%', pitch='+0Hz', voice_rate='+0%', voice_role='No', copysrt_rawvideo=False, clear_cache=True, translate_type=4, tts_type=0, recogn_type=0, model_name='large-v3-turbo', split_type=0, voice_autorate=True, video_autorate=False, cuda=True, name='I:/账号运营工作/BaiduSyncdisk/视频待投递/Granate旅行生活/素材/Granate Travelling walking indoor, #Hallstatt hotel, shorts, thigh high boots, high heels.mp4', basename='Granate Travelling walking indoor, #Hallstatt hotel, shorts, thigh high boots, high heels.mp4', ext='mp4', dirname='I:/账号运营工作/BaiduSyncdisk/视频待投递/Granate旅行生活/素材', shound_del_name=None, uuid='27d82477d4', remove_silent_mid=False, align_sub_audio=True, enable_diariz=False, nums_diariz=0, rephrase=1)


=
system:Windows-10-10.0.22621-SP0
version:v3.87
frozen:True
language:zh
root_dir:I:/fanyi/win-pyvideotrans-v3.87

1 Replies

这是典型的 GPU 显存不足错误。请按以下步骤解决:

  1. 降低模型精度:在菜单栏选择 工具/选项 -> 高级选项。在“Faster/OpenAI/Whisper 语音识别微调”区域,将“CUDA数据类型”从 float32 改为 int8float16。这是最有效的显存优化方法。
  2. 降低识别精度参数:在同一区域,将“识别准确度 (beam_size)”和“识别准确度 (best_of)”都改为 1。这会显著减少显存占用。
  3. 更换更小的模型:在主界面第2行“语音识别”中,将模型从 large-v3-turbo 更换为 mediumsmall
  4. 切换识别模式:在主界面第2行,将识别模式从“整体识别”改为“批量推理”。此模式对显存要求更低。
  5. 检查视频路径:您的视频路径包含中文字符和空格(如 I:/账号运营工作/...)。虽然软件支持,但为排除干扰,建议将视频文件复制到一个纯英文、无空格的短路径下(如 D:/video/test.mp4)再尝试处理。

操作顺序建议:先执行第1、2步,如果问题依旧再执行第3、4步。如果处理多个视频时出现此问题,还需在高级选项的“通用设置”中勾选“批量翻译时强制串行”。

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.

Donate: https://ko-fi.com/jianchang512

Trending Questions