Speech Resynthesis from Discrete Disentangled Self-Supervised Representations
This addresses speech synthesis and compression for applications like codecs, but it is incremental as it builds on existing self-supervised methods.
The paper tackles speech resynthesis by using self-supervised discrete representations to disentangle content, prosody, and speaker identity, achieving a rate of 365 bits per second with better quality than baselines.
We propose using self-supervised discrete representations for the task of speech resynthesis. To generate disentangled representation, we separately extract low-bitrate representations for speech content, prosodic information, and speaker identity. This allows to synthesize speech in a controllable manner. We analyze various state-of-the-art, self-supervised representation learning methods and shed light on the advantages of each method while considering reconstruction quality and disentanglement properties. Specifically, we evaluate the F0 reconstruction, speaker identification performance (for both resynthesis and voice conversion), recordings' intelligibility, and overall quality using subjective human evaluation. Lastly, we demonstrate how these representations can be used for an ultra-lightweight speech codec. Using the obtained representations, we can get to a rate of 365 bits per second while providing better speech quality than the baseline methods. Audio samples can be found under the following link: speechbot.github.io/resynthesis.