Skip to content
Tech · Oakland

Block releases Dessa, a voice model that runs locally and talks while it listens

Jack Dorsey and Ragavan Thurairatnam posted the model on Friday. Block said it was trained on human speech and one Nvidia GPU rack.

·
The Oakland marina
The marina in Oakland, California.

Block showed Dessa, a voice model that runs locally, in posts from Jack Dorsey and Ragavan Thurairatnam on Friday, October 23, TechCrunch reported. Block said Dessa was trained fully on human speech and on a single Nvidia GPU rack. The company described it as full duplex, meaning it can talk and listen at the same time, and said its knowledge comes from conversational data rather than a separate text pipeline.

Speech models are often paired with a separate language model for knowledge. Block said Dessa's knowledge comes from the conversational data it was trained on. The company said it built Dessa because voice is the most natural way to interact with an intelligent system.

From ARQS Sponsored

For B.C. teams putting stories like this into practice, ARQS offers Design Workbench, TradeHound, Website audit and Flow automation. In the ARQS Store, Wine Cellar Ledger — 1 Month Access, ARQS ContractDiff Desk, SponsorKit — Creator Media Kit & Rate Card Builder are among the newest tools available.