DiffNorm: Self-Supervised Normalization for Non-autoregressive Speech-to-speech Translation

Devs

DiffNorm: Self-Supervised Normalization for Non-autoregressive Speech-to-speech Translation | Read Paper on Bytez