Block Unveils Dessa, a Local Voice Model That Sounds Human

Block, the company led by Jack Dorsey, has released Dessa, a new voice model from its research lab that runs locally and reproduces human expressions such as laughs, pauses and cries. According to the company, the model was trained entirely on human speech using a single NVIDIA GPU.

Dessa is a full-duplex model, meaning it can listen and talk at the same time, similar to the GPT-Live model OpenAI released earlier this year. Voice systems are usually paired with a large language model that supplies knowledge. Block says Dessa instead draws its knowledge from the conversational data it was trained on.

Emotion built in, not tagged

Block argues that many current voice systems sound unnatural because text sits in the pipeline and drives speech generation. Models such as ElevenLabs v4 or Fish Audio Drama 3 rely on voice tags to steer emotion. Block claims emotion is built into Dessa itself, though it is unclear whether any kind of tagging guides it behind the scenes.

"Interacting with current voice AI can just plain suck. It turns your voice into text and throws away an incredible amount of information text alone can’t carry," the company said in a blog post.

Block's researchers shared several demo conversations in which the agent used different voices, cadences and emotions. In one, the bot said it was tired from work and began to sound tired. In another, it sounded emotional and almost broke down. A further clip showed a Block team member playing a guessing game with the assistant.

Local, but not yet public

The company says Dessa can run on consumer hardware such as a MacBook M4 Max or better. There is no public access at the moment, and Block has called for safety researchers to test the model, likely to probe cases where a voice agent could deceive the person on the other end.

The release matters for Block's business. The company sells financial tools to small businesses, and Square already uses voice AI to handle ordering. A proprietary model that runs on-device could power more business and consumer experiences at lower cost.