Towards Unsupervised Speech Recognition and Synthesis with Quantized Speech Representation Learning | Read Paper on Bytez