TOWARDS END-TO-END UNSUPERVISED SPEECH RECOGNITION

Unsupervised speech recognition has shown great potential to make Automatic Speech Recognition (ASR) systems accessible to every language. However, existing methods still heavily rely on hand-crafted pre-processing. Similar to the trend of making supervised speech recognition end-to-end, we introduce wav2vec-U 2.0 which does away with all audio-side preprocessing and improves accuracy through better architecture. In addition, we introduce an auxiliary self-supervised objective that ties model predictions back to the input. Experiments show that wav2vec-U 2.0 improves unsupervised recognition results across different languages while being conceptually simpler.

Automated summary

Unsupervised speech recognition has the potential to improve Automatic Speech Recognition (ASR) systems for all languages by eliminating pre-processing steps and introducing a self-supervised objective. The wav2vec-U 2.0 method shows improved results in unsupervised recognition across different languages while being conceptually simpler.