Block’s new voice model sounds eerily human

Block introduced Dessa, a speech-to-speech AI model trained on human speech that can listen and speak simultaneously and run locally on a MacBook Pro with an M4 Max or better. Block has not announced a public release date and is giving an early research preview to a small group of AI safety researchers.

• Block says Dessa was trained on human speech and can process audio directly while listening and speaking.· 2 sources
• The company says the model can run locally on a MacBook Pro with an M4 Max chip or better.· 2 sources
• Dessa is not publicly available; Block is offering early access to a small group of AI safety researchers and has not disclosed a public release date.· 2 sources