Block Releases Dessa Voice Model That Runs Locally and Mimics Human Emotion

Dessa is a full-duplex model, meaning it can talk and listen at the same time, like OpenAI's GPT-Live model released earlier this year, according to TechCrunch AI. Speech models are typically paired with a large language model for knowledge retrieval. Block said Dessa's knowledge stems from the conversational data it has.

Block argued that current voice models can sound unnatural and annoying because they use text in the pipeline to drive speech generation. Many models, such as ElevenLabs v4 or Fish Audio Drama 3, use voice tags to steer emotion in generated speech, according to the report. Block claims emotion is built into Dessa. It is not clear whether that emotion is being steered by some kind of tags.

"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 most powerful AIs in the world can't hear how you feel. We think that's ridiculous, leads to a worse experience, and is potentially dangerous," the company said in a blog post.

Block researchers posted multiple examples of conversations with Dessa in which the agent used different voices, cadences and emotions. In one conversation, the bot said it was tired from work and sounded tired at that point. In another, the bot spoke about a person making it comfortable while sounding emotional and almost breaking down. A further example showed a Block team member playing a game with the voice assistant.

The company said Dessa can run locally on a consumer-grade computer such as a MacBook M4 Max or better. There is no public access to the model at the moment. Block has issued a call for safety researchers to test Dessa, a step likely to examine cases in which a voice agent could trick people on the other end of a conversation.

Block sells financial solutions to small businesses, and Square already uses voice AI to handle ordering. According to TechCrunch AI, an on-device voice model of Block's own could power more business and consumer experiences at a lower cost.

Editor's Summary

Block has released Dessa, a voice model that runs locally and is designed to speak and listen simultaneously while expressing emotions such as tiredness and distress. The company has not made the model public and is inviting safety researchers to test it. Block's existing Square business already uses voice AI for ordering, and an on-device model could support more services at lower cost.