← Back to Model Beat
Models·2d ago·all news from November 11, 2026

Fish Audio Releases Live 3, a Full-Duplex Voice Model That Reacts in 186 ms

Fish Audio has launched Live 3, a full-duplex voice model capable of processing speech with a latency of 186 milliseconds. The system is designed to allow voice agents to listen while speaking and distinguish between conversational fillers and actual interruptions. The current release is restricted to English.

Covered by 1 source

Related stories

ModelsIntroducing AlphaProof Code: a breakthrough in verifying complex software at depthNov 10 · 5 sourcesModelsGemini 4 Carbon is now availableNov 10 · 7 sourcesModelsGoogle Search gets a major upgrade with Gemini 4 HydrogenNov 12 · 7 sourcesModelsGemini Decisions API: Fast, calibrated multimodal decisionsNov 11 · 6 sources