Closed lmmx closed 3 years ago
Now that the website pipeline is set up and before fine-tuning, switch the base model default for the large one (and confirm that it is possible to run model inference in a reasonable time, with the same segmentation level?)
Simply change the model name to facebook/wav2vec2-large-960h-lv60-self (https://huggingface.co/facebook/wav2vec2-large-960h-lv60-self)
facebook/wav2vec2-large-960h-lv60-self
https://github.com/lmmx/tap/blob/69ad95bedadad8d08bbb2ab8a6dc42bb577ef94c/src/tap/stt/transcribe.py#L7
Now that the website pipeline is set up and before fine-tuning, switch the base model default for the large one (and confirm that it is possible to run model inference in a reasonable time, with the same segmentation level?)
Simply change the model name to
facebook/wav2vec2-large-960h-lv60-self(https://huggingface.co/facebook/wav2vec2-large-960h-lv60-self)https://github.com/lmmx/tap/blob/69ad95bedadad8d08bbb2ab8a6dc42bb577ef94c/src/tap/stt/transcribe.py#L7