Topic

Speech-to-speech models

Models that take audio in and return audio out without a separate text stage, built for live spoken interaction rather than transcription.

Current clusters

build1 publisher

Gemini 3.8 Live Extended Thinking Can Send turnComplete While the Model Is Still Reasoning

Google's September 15 release pairs a fast speech-to-speech model with one that narrates its own reasoning aloud while tools run. Clients now have to read interactionStatus to know when a turn is actually over.

Publishers:dev.to

Reality

Evidence30
Adoption20
Hype gap+12
Incentives60
Confidence35