🔧 Connection Config
🎯 Model Selection
URL parameter 'model' — selects the upstream end-to-end engine
Override the server-side default voice
ASR transcription language (ISO 639-1); only takes effect for Audio-OG-realtime-*
🎙️ Upstream Audio
Browser demo only supports PCM upstream
🔊 Downstream Audio
Opus requires WebCodecs; this demo only supports PCM
🧠 Session
📡 Real-time Streams
🎧 ASR Transcription 0 utterances
💬 Assistant Reply 0 utterances
🔊 TTS Audio Stats
Total bytes0
Received delta packets0
Upstream bytes0