This model is a distilled version of the wav2vec2 model (https://arxiv.org/pdf/2006.11477.pdf). This model is 45% times smaller and twice as fast as the original wav2vec2 base model.

This model achieves the following results (speed is mesured for a batch size of 64):

notebook (executes seamlessly on google colab) at https://github.com/OthmaneJ/distil-wav2vec2