HybridCodec Boosts Speech LLMs by Marrying Discrete Tokens with Continuous Residuals
A new codec blends compressed discrete audio tokens with low‑dimensional continuous residuals, letting LLMs run autoregressive steps on the discrete side while upsampling continuous details later. This hybrid approach preserves speaker traits far better than pure tokenization and cuts the required autoregressive steps, making speech‑language models cheaper and higher‑fidelity.