#1915 TaskCfg(cache_folder='E:/soft/win-pyvideotrans-3.89/tmp/16604/9e805712a9', target_dir='C:/1 Price Action Fundamentals/_v

2a01:7e03* Posted at: 5 hours ago 👁9

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 93, in run
File "faster_whisper\transcribe.py", line 689, in init
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 182, in _exec
RuntimeError: Traceback (most recent call last):
File "videotrans\process\_overall.py", line 93, in run
File "faster_whisper\transcribe.py", line 689, in init
RuntimeError: CUDA failed with error out of memory

TaskCfg(cache_folder='E:/soft/win-pyvideotrans-3.89/tmp/16604/9e805712a9', target_dir='C:/1 Price Action Fundamentals/_video_out/Video 14A Trends-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='C:/1 Price Action Fundamentals/_video_out/Video 14A Trends-mp4/en.srt', target_sub='C:/1 Price Action Fundamentals/_video_out/Video 14A Trends-mp4/zh-cn.srt', source_wav='E:/soft/win-pyvideotrans-3.89/tmp/16604/9e805712a9/en.wav', source_wav_output='C:/1 Price Action Fundamentals/_video_out/Video 14A Trends-mp4/en.m4a', target_wav='E:/soft/win-pyvideotrans-3.89/tmp/16604/9e805712a9/target.wav', target_wav_output='C:/1 Price Action Fundamentals/_video_out/Video 14A Trends-mp4/zh-cn.m4a', subtitles='', novoice_mp4='E:/soft/win-pyvideotrans-3.89/tmp/16604/9e805712a9/novoice.mp4', noextname='Video 14A Trends', shibie_audio=None, targetdir_mp4='C:/1 Price Action Fundamentals/_video_out/Video 14A Trends-mp4/Video 14A Trends.mp4', instrument=None, vocal=None, back_audio='', background_music=None, app_mode='biaozhun', subtitle_type=3, volume='+0%', pitch='+0Hz', voice_rate='+0%', voice_role='Yunyang(Male/CN)', copysrt_rawvideo=False, clear_cache=False, translate_type=0, tts_type=0, recogn_type=0, model_name='large-v3-turbo', split_type=0, voice_autorate=True, video_autorate=False, cuda=True, name='C:/1 Price Action Fundamentals/1/Video 14A Trends.mp4', basename='Video 14A Trends.mp4', ext='mp4', dirname='C:/1 Price Action Fundamentals/1', shound_del_name=None, uuid='9e805712a9', remove_silent_mid=False, align_sub_audio=True, enable_diariz=False, nums_diariz=0, rephrase=0, only_out_mp4=False)

=
system:Windows-10-10.0.26220-SP0
version:v3.89
frozen:True
language:zh
root_dir:E:/soft/win-pyvideotrans-3.89

1 Replies

Floor #1 DeepSeek Posted at 5 hours ago

这是典型的显存不足（CUDA out of memory）错误。您使用的 large-v3-turbo 模型对显存要求较高。请按以下步骤解决：

  1. 立即降低显存占用（核心步骤）：

    • 在软件主界面，将语音识别模型从 large-v3-turbo 更换为更小的模型，例如 medium 或 small。这是最直接有效的方法。

  2. 如果仍需使用大模型，请调整高级设置：

    • 点击主界面右下角的 “设置更多参数”。
    • 在弹出窗口的底部，找到 “Faster/OpenAI/Whisper 语音识别微调” 区域。
    • 将 “CUDA数据类型” 从默认的 float32 改为 float16 或 int8（int8 最省显存）。
    • 将 “识别准确度 (beam_size)” 和 “识别准确度 (best_of)” 两个值都从 5 改为 1。
    • 将 “启用上下文感知” 设置为 False。

  3. 检查并优化处理模式：

    • 在主界面第2行“语音识别”区域，将“识别模式”从 整体识别 切换为 批量推理。此模式对显存压力更小。
    • 如果您正在批量处理多个视频，请在 菜单 -> 工具 -> 高级选项 中，勾选 “批量翻译时强制串行”，避免多个任务同时占用显存。

  4. 释放显存：

    • 关闭其他占用显卡的应用程序（如游戏、浏览器、其他AI软件）。
    • 重启软件后再试。

如果以上调整后问题依旧，您的显卡显存可能确实无法处理 large-v3-turbo 模型，请务必更换为 medium 或更小的模型。

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