#4476 TaskCfgSTT(is_cuda=True, uuid='012f25c124', cache_folder='D:/win-pyvideotrans-v3.99-428/tmp/12404/012f25c124', target_di

223.105* Posted at: 7 hours ago 👁17

语音识别阶段出错 [faster-whisper(本地)] Traceback (most recent call last):
File "videotrans\process\stt_fun.py", line 471, in faster_whisper
File "faster_whisper\transcribe.py", line 1851, in restore_speech_timestamps
File "faster_whisper\transcribe.py", line 1279, in generate_segments
File "faster_whisper\transcribe.py", line 1590, in add_word_timestamps
File "faster_whisper\transcribe.py", line 1709, in find_alignment
RuntimeError: cuBLAS failed with status CUBLAS_STATUS_NOT_SUPPORTED

Traceback (most recent call last):
File "videotrans\task\job.py", line 105, in run
File "videotrans\task\_speech2text.py", line 159, in recogn
File "videotrans\recognition\__init__.py", line 293, 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 289, in _new_process
RuntimeError: Traceback (most recent call last):
File "videotrans\process\stt_fun.py", line 471, in faster_whisper
File "faster_whisper\transcribe.py", line 1851, in restore_speech_timestamps
File "faster_whisper\transcribe.py", line 1279, in generate_segments
File "faster_whisper\transcribe.py", line 1590, in add_word_timestamps
File "faster_whisper\transcribe.py", line 1709, in find_alignment
RuntimeError: cuBLAS failed with status CUBLAS_STATUS_NOT_SUPPORTED
TaskCfgSTT(is_cuda=True, uuid='012f25c124', cache_folder='D:/win-pyvideotrans-v3.99-428/tmp/12404/012f25c124', target_dir='D:/win-pyvideotrans-v3.99-428/output/recogn', source_language=None, source_language_code=None, source_sub=None, source_wav=None, source_wav_output=None, target_language=None, target_language_code=None, target_sub='D:/win-pyvideotrans-v3.99-428/output/recogn/Jos Goossens 30 Years of Racing Pigeon Passion and Long-Distance.srt', target_wav=None, target_wav_output=None, name='D:/Jos Goossens 30 Years of Racing Pigeon Passion and Long-Distance.mp4', noextname='Jos Goossens 30 Years of Racing Pigeon Passion and Long-Distance', basename='Jos Goossens 30 Years of Racing Pigeon Passion and Long-Distance.mp4', ext='mp4', dirname='D:/', shound_del_name=None, detect_language='auto', recogn_type=0, model_name='large-v3', shibie_audio='D:/win-pyvideotrans-v3.99-428/tmp/12404/012f25c124/Jos Goossens 30 Years of Racing Pigeon Passion and Long-Distance-1778227267.2472885.wav', remove_noise=False, enable_diariz=False, nums_diariz=0, rephrase=0, fix_punc=False)


=
system:Windows-10-10.0.19045-SP0
version:v3.99
frozen:True
language:zh
root_dir:D:/win-pyvideotrans-v3.99-428

1 Replies

这是由于 CUDA 版本或 cuDNN 与你的 NVIDIA 显卡不兼容,导致 cuBLAS 无法正常调用。常见于新显卡(如 RTX 50 系)或驱动较旧的情况。

解决方案

  1. 升级 CUDA 和 cuDNN:确保已安装 CUDA 12.8 或更高版本 以及 cuDNN 9.11 或更高版本。前往 NVIDIA 官网下载对应版本,并确认环境变量包含 CUDA 的 binlib 路径。
  2. 针对 RTX 50 系显卡:请参考官方专题文档进行特殊配置(见下方链接)。
  3. 临时使用 CPU 处理:在主界面取消勾选 “CUDA 加速”,将改为使用 CPU 完成语音识别(速度会显著变慢)。
  4. 若以上无效,可尝试在 菜单 → 工具 → 高级选项 中,将 CUDA 数据类型 改为 int8,并降低 beam_sizebest_of 的值。

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