AI
Block builds a voice model that never learned to read
Jack Dorsey's Block has shown Dessa, a speech-to-speech model trained 'from scratch on nothing but human conversation' rather than on transcripts. It runs locally on a MacBook Pro M4 Max or better, and the company says it keeps the timing and emotional cues that text-first voice assistants flatten. Block published sample conversations, including one where Dessa mistakes the speaker for someone called Ronaldo, and will give a small group of safety researchers early access.