SLM-SS: Token-Based Speech Separation with Speech Language Models
SLM-SS is proposed, a token-based generative speech separation framework that models multi-talker mixtures with speech language models and introduces a per-talker non-autoregressive strategy that simplifies high-order modeling and improves the stability.
Chen-Da Li, Tian-Hua Li, Yi-Ming Chen et al.
· 0 citations