Gemini 3.1 Flash Live
A native audio large model introduced by Google DeepMind that supports real-time voice conversation and multilingual understanding and generation without an ASR/TTS cascade, characterized by low latency and the preservation of paralinguistic information.
Timeline (last 90 days)
Gemini 3.1 Flash Live supports about 90 languages during the preview phase.
Native audio models (such as Gemini 3.1 Flash Live) understand and generate directly in the audio token space, eliminating intermediate transcription, preserving speaker emotion and speech-rate characteristics, and significantly reducing latency.
All Facts (2)
Gemini 3.1 Flash Live supports about 90 languages during the preview phase.
50%UnverifiedNative audio models (such as Gemini 3.1 Flash Live) understand and generate directly in the audio token space, eliminating intermediate transcription, preserving speaker emotion and speech-rate characteristics, and significantly reducing latency.
50%