Polish ASR: Fine-Tuning Wav2Vec2

Hi, I’m currently training PL on CommonVoice on colab.
The model should be good enough, but it’s also interesting to see how it would behave with Multilingual LibriSpeech added to the dataset. I won’t have time nor computational power to check that so feel free to proceed :slight_smile:

My model will be available on mbien/wav2vec2-large-xlsr-polish after the training, I’ll also post the resulting metrics here!

1 Like