Block's AI research lab, led by Jack Dorsey's fintech company, released a research preview of Dessa, a full-duplex voice conversation model trained exclusively on human conversational speech. Unlike conventional models, Dessa was built from scratch without text data or pretrained foundations, enabling it to listen and speak simultaneously.
The model was trained using a single NVIDIA GB200 NVL72 GPU and can run locally on a MacBook Pro M4 Max or higher at faster-than-real-time speed. Project lead Ragavan Thurairatnam described the release as "the GPT-3 moment for voice." Dessa is not yet broadly available; Block has published recorded conversation demos and granted early access to a small group of AI safety researchers.
Block Releases Dessa, a Voice-Only Conversational AI Model That Runs Locally on MacBook
Disclaimer: The content provided on Phemex News is for informational purposes only. We do not guarantee the quality, accuracy, or completeness of the information sourced from third-party articles. The content on this page does not constitute financial or investment advice. We strongly encourage you to conduct you own research and consult with a qualified financial advisor before making any investment decisions.
