Gemini Omni
Kembali ke semua artikel
10 min baca

Panduan Gemini 3.1 Flash Live API: ejen suara, kamera dan perkongsian skrin masa nyata

Bina ejen multimodal latensi rendah dengan Gemini 3.1 Flash Live — WebSocket, audio natif, kamera/skrin 1 FPS, token efemeral.

Gemini Live APIGemini 3.1 Flash LiveReal-TimeVoice AgentsMultimodalDevelopers2026Bahasa Melayu

Mengapa Gemini 3.1 Flash Live penting Ogos 2026

Google melancarkan Gemini 3.1 Flash Live melalui Gemini Live API Mac 2026 — Ogos menjadi model disyorkan untuk ejen suara+visi produksi. Live mengekalkan sesi WebSocket kekal tanpa STT/TTS berasingan.

Ogos 2026: 2.5 Flash Live deprecated; ekosistem matang; Live + Gemini 3.7 Flash (pengumuman 3.7 Flash).

Architecture: WebSocket sessions

Live API is stateful — persistent WSS bidirectional streams for audio, JPEG/PNG video (≤1 FPS), and text. First message: BidiGenerateContentSetup; then BidiGenerateContentRealtimeInput.

LayerRole
TransportStateful WSS
InputPCM16 @ 16 kHz, video ≤1 FPS
OutputPCM @ 24 kHz, transcripts, tools

Server-to-server (proxy) or client-to-server (ephemeral tokens required).

Native audio pipeline

Input PCM16 16 kHz; output 24 kHz. Browser float32 → PCM16. VAD for turns; manual activityStart/activityEnd for push-to-talk.

Camera and screen share

1 FPS JPEG/PNG — camera Q&A, screen code review. Default TURN_INCLUDES_AUDIO_ACTIVITY_AND_ALL_VIDEO — send frames only during speech to reduce cost. Limits: 15 min audio-only, 2 min audio+video.

Tools, thinkingLevel, security

Sync tool calling + Search grounding. thinkingLevel: minimalhigh. 90+ languages. Ephemeral tokens — never embed API keys in clients.

Migration from 2.5

Use gemini-3.1-flash-live-preview; thinkingBudgetthinkingLevel; send_realtime_input for text; multi-part events; no NON_BLOCKING tools.

3.1 Live vs 3.7 REST

Live: WebSocket, sub-second voice, 128K context. 3.7: HTTP, coding/RAG, 2.5M tokens.