#4470 TaskCfgVTT(is_cuda=True, uuid='af46875c30', cache_folder='E:/usr/Application/AI-Models/apps/pyvideotrans/tmp/28948/af468

141.164* Posted at: 23 hours ago 👁39

语音识别阶段出错 [faster-whisper(本地)] Traceback (most recent call last):
File "E:\usr\Application\AI-Models\apps\pyvideotrans\videotrans\process\stt_fun.py", line 390, in faster_whisper

segments, info = batched_model.transcribe(
                 ^^^^^^^^^^^^^^^^^^^^^^^^^

File "E:\usr\Application\AI-Models\apps\pyvideotrans\venv\Lib\site-packages\faster_whisper\transcribe.py", line 515, in transcribe

np.stack([pad_or_trim(feature) for feature in features]) if features else []
          ^^^^^^^^^^^^^^^^^^^^

File "E:\usr\Application\AI-Models\apps\pyvideotrans\venv\Lib\site-packages\faster_whisper\audio.py", line 121, in pad_or_trim

array = np.pad(array, pad_widths)
        ^^^^^^^^^^^^^^^^^^^^^^^^^

File "E:\usr\Application\AI-Models\apps\pyvideotrans\venv\Lib\site-packages umpy\lib\arraypad.py", line 798, in pad

padded, original_area_slice = _pad_simple(array, pad_width)
                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "E:\us
......
ideotrans\videotrans\recognition\_overall.py", line 105, in _faster

raws=self._new_process(callback=faster_whisper,title=title,is_cuda=self.is_cuda,kwargs=kwargs)
     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "E:\usr\Application\AI-Models\apps\pyvideotrans\videotrans\configure\_base.py", line 289, in _new_process

raise RuntimeError(err)

RuntimeError: Traceback (most recent call last):
File "E:\usr\Application\AI-Models\apps\pyvideotrans\videotrans\process\stt_fun.py", line 390, in faster_whisper

segments, info = batched_model.transcribe(
                 ^^^^^^^^^^^^^^^^^^^^^^^^^

File "E:\usr\Application\AI-Models\apps\pyvideotrans\venv\Lib\site-packages\faster_whisper\transcribe.py", line 515, in transcribe

np.stack([pad_or_trim(feature) for feature in features]) if features else []
          ^^^^^^^^^^^^^^^^^^^^

File "E:\usr\Application\AI-Models\apps\pyvideotrans\venv\Lib\site-packages\faster_whisper\audio.py", line 121, in pad_or_trim

array = np.pad(array, pad_widths)
        ^^^^^^^^^^^^^^^^^^^^^^^^^

File "E:\usr\Application\AI-Models\apps\pyvideotrans\venv\Lib\site-packages umpy\lib\arraypad.py", line 798, in pad

padded, original_area_slice = _pad_simple(array, pad_width)
                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "E:\usr\Application\AI-Models\apps\pyvideotrans\venv\Lib\site-packages umpy\lib\arraypad.py", line 114, in _pad_simple

padded = np.empty(new_shape, dtype=array.dtype, order=order)
         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

numpy.core._exceptions._ArrayMemoryError: Unable to allocate 1.46 MiB for an array with shape (128, 3000) and data type float32
TaskCfgVTT(is_cuda=True, uuid='af46875c30', cache_folder='E:/usr/Application/AI-Models/apps/pyvideotrans/tmp/28948/af46875c30', target_dir='G:/[Work-Space]/[Screen-Rec]/[Video-Cache]/_video_out/2026-05-03_23-09-27-mp4', source_language='英语', source_language_code='en', source_sub='G:/[Work-Space]/[Screen-Rec]/[Video-Cache]/_video_out/2026-05-03_23-09-27-mp4/en.srt', source_wav='E:/usr/Application/AI-Models/apps/pyvideotrans/tmp/28948/af46875c30/en.wav', source_wav_output='G:/[Work-Space]/[Screen-Rec]/[Video-Cache]/_video_out/2026-05-03_23-09-27-mp4/en.m4a', target_language='简体中文', target_language_code='zh-cn', target_sub='G:/[Work-Space]/[Screen-Rec]/[Video-Cache]/_video_out/2026-05-03_23-09-27-mp4/zh-cn.srt', target_wav='E:/usr/Application/AI-Models/apps/pyvideotrans/tmp/28948/af46875c30/target.wav', target_wav_output='G:/[Work-Space]/[Screen-Rec]/[Video-Cache]/_video_out/2026-05-03_23-09-27-mp4/zh-cn.m4a', name='G:/[Work-Space]/[Screen-Rec]/[Video-Cache]/2026-05-03_23-09-27.mp4', noextname='2026-05-03_23-09-27', basename='2026-05-03_23-09-27.mp4', ext='mp4', dirname='//Epa-00/30tb/[Work-Space]/[Screen-Rec]/[Video-Cache]', shound_del_name=None, translate_type=1, 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='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=True, subtitle_language=None, app_mode='tiqu', subtitles='', targetdir_mp4='G:/[Work-Space]/[Screen-Rec]/[Video-Cache]/_video_out/2026-05-03_23-09-27-mp4/2026-05-03_23-09-27.mp4', novoice_mp4='E:/usr/Application/AI-Models/apps/pyvideotrans/tmp/28948/af46875c30/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=False, output_srt=1, copysrt_rawvideo=True)


=
system:Windows-10-10.0.19045-SP0
version:v3.98
frozen:False
language:zh
root_dir:E:/usr/Application/AI-Models/apps/pyvideotrans
Python: 3.12.12 | packaged by Anaconda, Inc. | (main, Oct 21 2025, 20:05:38) [MSC v.1929 64 bit (AMD64)]

1 Replies

这是因内存分配失败导致的错误,通常由系统内存不足、内存碎片或模型/参数配置过高引起。您使用的 large-v3-turbo 模型和 float32 数据类型可能超出了当前可用资源。

请按以下步骤尝试解决:

  1. 更换为更小的模型:在主界面语音识别渠道右侧模型下拉中选择 smallbase,这两个模型内存占用远小于 large-v3-turbo
  2. 降低精度与并发

    • 打开 菜单 → 工具/选项 → 高级选项 → 语音识别参数
    • CUDA数据类型 改为 float16int8
    • beam_sizebest_of 均改为 1
    • 禁用 启用上下文感知
  3. 启用音频预分割:在同一高级设置中,勾选 Whisper预分割音频,可将长音频切分后再识别,显著降低单次内存分配压力。
  4. 释放系统内存:关闭其他占用内存的程序,重启软件后重试。如果长期未重启系统,建议重启计算机后仅运行本软件。

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