Block unveils Dessa full-duplex speech model trained on conversation audio
Original titleBlock Unveils Dessa, a Full-Duplex Speech Model Trained Only on Speech
- Outlet
- MarkTechPost
- Published
- Importance
- 4/5
Summary
Block's AI lab released Dessa, a full-duplex speech-to-speech model trained exclusively on human conversation audio that runs locally on a MacBook Pro M4 Max.
Original excerpt
Block’s AI lab released Dessa, a full-duplex speech-to-speech model trained only on human conversation audio. It runs locally on a MacBook Pro M4 Max, but Block has published no size, benchmarks or weights. The post Block Unveils Dessa, a Full-Duplex Speech Model Trained Only on Speech appeared first on MarkTechPost.
#block#speech#dessa