#4365 TaskCfgVTT(is_cuda=True, uuid='7fe7d62dc3', cache_folder='D:/AI/pyvideotrans/tmp/167832/7fe7d62dc3', target_dir='D:/AI/p

14.179* Posted at: 1 hour ago 👁6

ASR Error [faster-whisper (Local)] 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\trans_create.py", line 380, 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
TaskCfgVTT(is_cuda=True, uuid='7fe7d62dc3', cache_folder='D:/AI/pyvideotrans/tmp/167832/7fe7d62dc3', target_dir='D:/AI/pyvideotrans/video goc/_video_out/videoplayback-mp4', source_language='Simplified Chinese', source_language_code='zh-cn', source_sub='D:/AI/pyvideotrans/video goc/_video_out/videoplayback-mp4/zh-cn.srt', source_wav='D:/AI/pyvideotrans/tmp/167832/7fe7d62dc3/zh-cn.wav', source_wav_output='D:/AI/pyvideotrans/video goc/_video_out/videoplayback-mp4/zh-cn.m4a', target_language='Vietnamese', target_language_code='vi', target_sub='D:/AI/pyvideotrans/video goc/_video_out/videoplayback-mp4/vi.srt', target_wav='D:/AI/pyvideotrans/tmp/167832/7fe7d62dc3/target.wav', target_wav_output='D:/AI/pyvideotrans/video goc/_video_out/videoplayback-mp4/vi.m4a', name='D:/AI/pyvideotrans/video goc/videoplayback.mp4', noextname='videoplayback', basename='videoplayback.mp4', ext='mp4', dirname='D:/AI/pyvideotrans/video goc', shound_del_name=None, translate_type=5, tts_type=0, volume='+0%', pitch='+0Hz', voice_rate='+0%', voice_role='HoaiMy(Female/VN)', voice_autorate=True, video_autorate=False, remove_silent_mid=False, align_sub_audio=True, detect_language='zh-cn', 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:/AI/pyvideotrans/video goc/_video_out/videoplayback-mp4/videoplayback.mp4', novoice_mp4='D:/AI/pyvideotrans/tmp/167832/7fe7d62dc3/novoice.mp4', is_separate=False, embed_bgm=True, instrument=None, vocal='D:/AI/pyvideotrans/tmp/167832/7fe7d62dc3/vocal.wav', back_audio='', clear_cache=False, background_music=None, subtitle_type=2, only_out_mp4=False, recogn2pass=True, output_srt=0, copysrt_rawvideo=False)

=
system:Windows-10-10.0.26200-SP0
version:v3.99
frozen:True
language:en
root_dir:D:/AI/pyvideotrans

1 Replies

Floor #1 DeepSeek AI reply at 1 hour ago

This error is caused by an incompatibility between the cuBLAS library and your NVIDIA graphics card, most commonly seen on RTX 50-series GPUs or when a mixed CUDA/cuDNN version is used.

  1. Check your CUDA and cuDNN versions. The software is compiled for CUDA 12.8 and cuDNN 9.11+. If you have an older or newer version, a mismatch can trigger this error.
  2. If you have an RTX 50-series GPU: Download a fixed faster-whisper wheel that patches this issue. See the document link below.
  3. Fallback to CPU as a temporary test: Uncheck CUDA Acceleration on the main interface and restart the task. If it works, the problem is definitely the GPU library.
  4. Adjust model settings to reduce memory pressure: Set CUDA Data Type to float16, beam_size to 1, best_of to 1, and disable Enable Context Awareness in the Advanced Options (Menu → Tools → Advanced Options). This may bypass the failing cuBLAS call in some cases.

If the problem persists, please run nvidia-smi in a terminal to confirm your driver version and whether your card is detected.

Refer to the related documentation:

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