delayed-streams-modeling

Project Description

Kyutai's Speech-To-Text and Text-To-Speech models based on the Delayed Streams Modeling framework.