#4711 TaskCfgVTT(is_cuda=True, uuid='25a1536ea2', cache_folder='D:/project/Videotrans/raw-source/tmp/17932/25a1536ea2', target

18.142* Posted at: 1 hour ago

语音识别阶段出错 [faster-whisper(本地)] Traceback (most recent call last):
File "D:\project\Videotrans\raw-source\videotrans\process\stt_fun.py", line 346, in faster_whisper

model = WhisperModel(

File "D:\project\Videotrans\raw-source.venv\lib\site-packages\faster_whisper\transcribe.py", line 689, in init

self.model = ctranslate2.models.Whisper(

ValueError: Requested float16 compute type, but the target device or backend do not support efficient float16 computation.

Traceback (most recent call last):
File "D:\project\Videotrans\raw-source\videotrans\configure\_base.py", line 297, in _new_process

_rs = future.result(timeout=timeout)

File "C:\Users\savior\AppData\Roaming\uv\python\cpython-3.10.19-windows-x86_64-none\lib\concurrent\futures\_base.py", line 458, in result

return self.__get_result()

File "C:\Users\savior\AppData\Roaming\uv\python\cpython-3.10.19-windows-x86_64-none\lib\concurrent\futures\_base.py", line 403, in __get_result

raise self._

......
ture was running or pending.

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "D:\project\Videotrans\raw-source\videotrans\task\job.py", line 105, in run

trk.recogn()

File "D:\project\Videotrans\raw-source\videotrans\task\trans_create.py", line 368, in recogn

raw_subtitles = run_recogn(

File "D:\project\Videotrans\raw-source\videotrans\recognition\__init__.py", line 292, in run

return FasterAll(**kwargs).run()

File "D:\project\Videotrans\raw-source\videotrans\recognition\_base.py", line 143, in run

res = self._exec()

File "D:\project\Videotrans\raw-source\videotrans\recognition\_overall.py", line 33, in _exec

raws = self._faster()

File "D:\project\Videotrans\raw-source\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 "D:\project\Videotrans\raw-source\videotrans\configure\_base.py", line 370, in _new_process

raise RuntimeError(err)

RuntimeError: Traceback (most recent call last):
File "D:\project\Videotrans\raw-source\videotrans\process\stt_fun.py", line 346, in faster_whisper

model = WhisperModel(

File "D:\project\Videotrans\raw-source.venv\lib\site-packages\faster_whisper\transcribe.py", line 689, in init

self.model = ctranslate2.models.Whisper(

ValueError: Requested float16 compute type, but the target device or backend do not support efficient float16 computation.
TaskCfgVTT(is_cuda=True, uuid='25a1536ea2', cache_folder='D:/project/Videotrans/raw-source/tmp/17932/25a1536ea2', target_dir='E:/1/_video_out/1 1. Introduction - What You Will Build In This Course-mp4', source_language='英语', source_language_code='en', source_sub='E:/1/_video_out/1 1. Introduction - What You Will Build In This Course-mp4/en.srt', source_wav='D:/project/Videotrans/raw-source/tmp/17932/25a1536ea2/en.wav', source_wav_output='E:/1/_video_out/1 1. Introduction - What You Will Build In This Course-mp4/en.m4a', target_language='简体中文', target_language_code='zh-cn', target_sub='E:/1/_video_out/1 1. Introduction - What You Will Build In This Course-mp4/zh-cn.srt', target_wav='D:/project/Videotrans/raw-source/tmp/17932/25a1536ea2/target.wav', target_wav_output='E:/1/_video_out/1 1. Introduction - What You Will Build In This Course-mp4/zh-cn.m4a', name='E:/1/Udemy - Build an AI-Powered SaaS with Next.js 16, ChatGPT API & RAG part1/1 1. Introduction - What You Will Build In This Course.mp4', noextname='1 1. Introduction - What You Will Build In This Course', basename='1 1. Introduction - What You Will Build In This Course.mp4', ext='mp4', dirname='E:/1/Udemy - Build an AI-Powered SaaS with Next.js 16, ChatGPT API & RAG part1', shound_del_name=None, translate_type=3, tts_type=34, volume='+0%', pitch='+0Hz', voice_rate='+0%', voice_role='白桦', voice_autorate=True, 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=1, fix_punc=False, subtitle_language=None, app_mode='biaozhun', subtitles='', targetdir_mp4='E:/1/_video_out/1 1. Introduction - What You Will Build In This Course-mp4/1 1. Introduction - What You Will Build In This Course.mp4', novoice_mp4='D:/project/Videotrans/raw-source/tmp/17932/25a1536ea2/novoice.mp4', is_separate=True, embed_bgm=True, instrument='D:/project/Videotrans/raw-source/tmp/17932/25a1536ea2/instrument.wav', vocal='D:/project/Videotrans/raw-source/tmp/17932/25a1536ea2/vocal.wav', back_audio='', clear_cache=False, background_music=None, subtitle_type=4, only_out_mp4=False, recogn2pass=True, output_srt=2, copysrt_rawvideo=False)


=
system:Windows-10-10.0.26200-SP0
version:v3.99
frozen:False
language:zh
root_dir:D:/project/Videotrans/raw-source
Python: 3.10.19 (main, Feb 12 2026, 00:37:08) [MSC v.1944 64 bit (AMD64)]

1 Replies

这是由于您的显卡或CUDA环境不支持高效的float16计算类型。faster-whisperctranslate2在启用CUDA时默认尝试使用float16,但您的显卡(如10系列或老旧)或CUDA版本未满足要求。

解决方案

  1. 进入高级选项:在软件顶部菜单栏点击 工具/选项 → 高级选项 → 语音识别参数
  2. 修改CUDA数据类型:将 CUDA数据类型 从默认的 float16 改为 int8(兼容性最好,速度损失小且节省显存),或改为 float32(速度慢但兼容)。
  3. 保存并重试:保存设置后,重新处理视频。
如果仍报错,请确认已安装CUDA 12.8+ 和 cuDNN 9.11+,并查看 nvcc -V 输出。若显卡不支持CUDA(如仅支持CUDA 11.x或更低),建议在软件主界面 取消勾选“CUDA加速” 并改用CPU处理。
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