insanely_fast_whisper_colab.ipynb doesn't work

[danielKlasss]

Hi,

I find your project very interesting, therefore I tried to run the demo notebook in a T4 runtime on colab.
But the execution of the cell !pipx run insanely-fast-whisper --file-name https://huggingface.co/datasets/reach-vb/random-audios/resolve/main/ted_60.wav fails for me:

/root/.cache/pipx/fbbfdc728a6642e/lib/python3.10/site-packages/pyannote/audio/core/io.py:43: UserWarning: torchaudio._backend.set_audio_backend has been deprecated. With dispatcher enabled, this function is no-op. You can remove the function call.
  torchaudio.set_audio_backend("soundfile")
/root/.cache/pipx/fbbfdc728a6642e/lib/python3.10/site-packages/torch_audiomentations/utils/io.py:27: UserWarning: torchaudio._backend.set_audio_backend has been deprecated. With dispatcher enabled, this function is no-op. You can remove the function call.
  torchaudio.set_audio_backend("soundfile")
config.json: 100% 1.27k/1.27k [00:00<00:00, 5.89MB/s]
model.safetensors: 100% 3.09G/3.09G [00:25<00:00, 119MB/s]
generation_config.json: 100% 3.87k/3.87k [00:00<00:00, 15.4MB/s]
tokenizer_config.json: 100% 283k/283k [00:00<00:00, 12.4MB/s]
vocab.json: 100% 1.04M/1.04M [00:00<00:00, 3.90MB/s]
tokenizer.json: 100% 2.48M/2.48M [00:00<00:00, 30.2MB/s]
merges.txt: 100% 494k/494k [00:00<00:00, 54.1MB/s]
normalizer.json: 100% 52.7k/52.7k [00:00<00:00, 48.9MB/s]
added_tokens.json: 100% 34.6k/34.6k [00:00<00:00, 93.6MB/s]
special_tokens_map.json: 100% 2.07k/2.07k [00:00<00:00, 9.21MB/s]
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
preprocessor_config.json: 100% 340/340 [00:00<00:00, 1.75MB/s]
Traceback (most recent call last):
  File "/root/.cache/pipx/fbbfdc728a6642e/bin/insanely-fast-whisper", line 8, in <module>
    sys.exit(main())
  File "/root/.cache/pipx/fbbfdc728a6642e/lib/python3.10/site-packages/insanely_fast_whisper/cli.py", line 105, in main
    pipe.model = pipe.model.to_bettertransformer()
  File "/root/.cache/pipx/fbbfdc728a6642e/lib/python3.10/site-packages/transformers/modeling_utils.py", line 4302, in to_bettertransformer
    return BetterTransformer.transform(self)
  File "/usr/lib/python3.10/contextlib.py", line 79, in inner
    return func(*args, **kwds)
  File "/root/.cache/pipx/fbbfdc728a6642e/lib/python3.10/site-packages/optimum/bettertransformer/transformation.py", line 228, in transform
    raise NotImplementedError(
NotImplementedError: The model type whisper is not yet supported to be used with BetterTransformer. Feel free to open an issue at https://github.com/huggingface/optimum/issues if you would like this model type to be supported. Currently supported models are: dict_keys(['albert', 'bark', 'bart', 'bert', 'bert-generation', 'blenderbot', 'bloom', 'camembert', 'blip-2', 'clip', 'codegen', 'data2vec-text', 'deit', 'distilbert', 'electra', 'ernie', 'fsmt', 'gpt2', 'gptj', 'gpt_neo', 'gpt_neox', 'hubert', 'layoutlm', 'm2m_100', 'marian', 'markuplm', 'mbart', 'opt', 'pegasus', 'rembert', 'prophetnet', 'roberta', 'roc_bert', 'roformer', 'splinter', 'tapas', 't5', 'vilt', 'vit', 'vit_mae', 'vit_msn', 'wav2vec2', 'xlm-roberta', 'yolos']).

Best wishes

Daniel

[afiaka87]

I've reported a similar issue running on an RTX 3060 gaming PC with linux here:

#133

[Vaibhavs10]

Hi @danielKlasss & @afiaka87 I just released a patch and this should be fixed now. The issue crept up due to an update in optimum: huggingface/optimum#1590

Make sure to get the most recent version of insanely-fast-whisper:

1. pipx uninstall insanely-fast-whisper
2. pipx install insanely-fast-whisper==0.0.13 --force

P.S. On the Colab it should work out of the box since we create a fresh virtualenv everytime it is run :)

[danielKlasss]

@Vaibhavs10 Thank you, I can confirm that the problem is fixed.