#4067 TaskCfgSTT(is_cuda=True, uuid='015483d275', cache_folder='E:/ytrans/tmp/3636/015483d275', target_dir='E:/ytrans/output/r

202.150* Posted at: 4 hours ago 👁9

ASR Error [faster-whisper (Local)] Traceback (most recent call last):
File "videotrans\process\stt_fun.py", line 257, in faster_whisper
File "faster_whisper\transcribe.py", line 586, in _batched_segments_generator
File "faster_whisper\transcribe.py", line 120, in forward
File "faster_whisper\transcribe.py", line 209, in generate_segment_batched
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 106, in run
File "videotrans\task\_speech2text.py", line 153, in recogn
File "videotrans\recognition\__init__.py", line 266, in run
File "videotrans\recognition\_base.py", line 142, in run
File "videotrans\recognition\_overall.py", line 63, in _exec
File "videotrans\recognition\_overall.py", line 136, in _faster
File "videotrans\configure\_base.py", line 290, in _new_process
RuntimeError: Traceback (most recent call last):
File "videotrans\process\stt_fun.py", line 257, in faster_whisper
File "faster_whisper\transcribe.py", line 586, in _batched_segments_generator
File "faster_whisper\transcribe.py", line 120, in forward
File "faster_whisper\transcribe.py", line 209, in generate_segment_batched
File "faster_whisper\transcribe.py", line 1400, in encode
RuntimeError: CUDA failed with error out of memory
TaskCfgSTT(is_cuda=True, uuid='015483d275', cache_folder='E:/ytrans/tmp/3636/015483d275', target_dir='E:/ytrans/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='E:/ytrans/output/recogn/05. セットアップ画面の機能.srt', target_wav=None, target_wav_output=None, name='E:/2023/personal/inspo2/Animation/Spine2D/Hanseryun/Videos/05. セットアップ画面の機能.mp4', noextname='05. セットアップ画面の機能', basename='05. セットアップ画面の機能.mp4', ext='mp4', dirname='E:/2023/personal/inspo2/Animation/Spine2D/Hanseryun/Videos', shound_del_name=None, detect_language='ko', recogn_type=0, model_name='large-v3-turbo', shibie_audio='E:/ytrans/tmp/3636/015483d275/05. セットアップ画面の機能-1776173261.8451676.wav', remove_noise=False, enable_diariz=False, nums_diariz=0, rephrase=1, fix_punc=False)

=
system:Windows-10-10.0.26200-SP0
version:v3.97
frozen:True
language:en
root_dir:E:/ytrans

1 Replies

Floor #1 DeepSeek AI reply at 4 hours ago

This error is caused by insufficient GPU memory (VRAM) when using the faster-whisper model with CUDA acceleration.

Solutions:

  1. Disable "Secondary Recognition": If you have "Secondary Recognition" enabled in the main interface (top-right corner), turn it off first. When processing multiple videos, secondary recognition from earlier tasks can run concurrently with the initial transcription of later tasks, potentially requiring 2-3 times the VRAM.
  2. Use a smaller model: The large-v3-turbo model requires significant VRAM. Switch to a smaller model like medium, small, or base in the "Speech Recognition" dropdown.

  3. Adjust advanced settings (Menu -> Tools/Options -> Advanced Options): In the "Speech Recognition Parameters" section, modify the following to reduce VRAM usage at the cost of some accuracy:

    • CUDA Data Type: Change from float32 to float16 or int8.
    • beam_size: Change from 5 to 1.
    • best_of: Change from 5 to 1.
    • Enable Context Awareness: Set to false.
  4. Check for multiple GPUs (v3.99+): If you have multiple NVIDIA GPUs, version v3.99 and later will automatically use the one with the most available VRAM by default. Consider upgrading to v3.99 if you haven't already.
  5. Ensure only one GPU task runs (Advanced Options): In "General Settings", set GPU Simultaneous Tasks [Restart Required] to 1 to prevent multiple tasks from overloading the GPU.

Recommended immediate actions: Disable "Secondary Recognition" and switch to a medium or small model. If the error persists, proceed with step 3 (adjusting advanced settings).

Please refer to the relevant 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