Block Introduces Dessa, a Speech Model Built for Live Conversation

Block has unveiled Dessa, a speech-to-speech AI model designed to listen and speak at the same time during live conversation. The company says it was trained from scratch on human conversation audio alone and can run locally on a MacBook Pro M4 Max or better. The model’s launch page describes an audio-in, audio-out system that can interrupt and use conversational backchannels while the other speaker is still talking.