search

Vaibhavs10 / insanely-fast-whisper

Apache License 2.0
7.5k stars 529 forks source link

Torch not compiled with CUDA enabled #124

Closed marsonal2023 closed 9 months ago

marsonal2023 commented 9 months ago

raise AssertionError("Torch not compiled with CUDA enabled") AssertionError: Torch not compiled with CUDA enabled ??

Vaibhavs10 commented 9 months ago

Hi @marsonal2023 - Can you provide more details for the error? What CLI command are you using? What's your system configuration and so on?

leoliu606 commented 9 months ago

(nf-whisper311) PS D:\ai\live-main\live-ts> insanely-fast-whisper --file-name https://alldogen.oss-cn-shanghai.aliyuncs.com/vd1106/a24591ce31ffb76fbbc28e1f34711e49/audio.webm Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained. Traceback (most recent call last): File "", line 198, in _run_module_as_main File "", line 88, in _run_code File "D:\app\conda\envs\nf-whisper311\Scripts\insanely-fast-whisper.exe__main.py", line 7, in File "D:\app\conda\envs\nf-whisper311\Lib\site-packages\insanely_fast_whisper\cli.py", line 78, in main pipe = pipeline( ^^^^^^^^^ File "D:\app\conda\envs\nf-whisper311\Lib\site-packages\transformers\pipelines__init__.py", line 1070, in pipeline return pipeline_class(model=model, framework=framework, task=task, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\app\conda\envs\nf-whisper311\Lib\site-packages\transformers\pipelines\automatic_speech_recognition.py", line 239, in init
self.model.to(device) File "D:\app\conda\envs\nf-whisper311\Lib\site-packages\transformers\modeling_utils.py", line 2460, in to return super().to(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\app\conda\envs\nf-whisper311\Lib\site-packages\torch\nn\modules\module.py", line 1160, in to return self._apply(convert) ^^^^^^^^^^^^^^^^^^^^ File "D:\app\conda\envs\nf-whisper311\Lib\site-packages\torch\nn\modules\module.py", line 810, in _apply module._apply(fn) File "D:\app\conda\envs\nf-whisper311\Lib\site-packages\torch\nn\modules\module.py", line 810, in _apply module._apply(fn) File "D:\app\conda\envs\nf-whisper311\Lib\site-packages\torch\nn\modules\module.py", line 810, in _apply module._apply(fn) File "D:\app\conda\envs\nf-whisper311\Lib\site-packages\torch\nn\modules\module.py", line 833, in _apply param_applied = fn(param) ^^^^^^^^^ File "D:\app\conda\envs\nf-whisper311\Lib\site-packages\torch\nn\modules\module.py", line 1158, in convert return t.to(device, dtype if t.is_floating_point() or t.is_complex() else None, non_blocking) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\app\conda\envs\nf-whisper311\Lib\site-packages\torch\cuda\init__.py", line 289, in _lazy_init raise AssertionError("Torch not compiled with CUDA enabled") AssertionError: Torch not compiled with CUDA enabled (nf-whisper311) PS D:\ai\live-main\live-ts>

wihndows , conda, python=3.11, nvidia4090,cuda118

@Vaibhavs10 Good

marsonal2023 commented 9 months ago

yes what is what excatly happened to me too. google search results told me to reinsall torch but i already have it installed

Vaibhavs10 commented 9 months ago

This looks like the same issue as the second question in the FAQ: https://github.com/Vaibhavs10/insanely-fast-whisper#frequently-asked-questions

This is a windows specific issue. The second question is known to fix this issue.

Sorry about this, packaging can be quite messy.

(feel free to re-open this issue if it doesn't fix)

IsaackRasmussen commented 5 months ago

For me, it helped to get the latest version by referencing the this repository when installing https://github.com/Vaibhavs10/insanely-fast-whisper/issues/213