Deepgram Nova-3 STT (WebSocket) → Groq LLM (openai/gpt-oss-20b) → Deepgram Flux TTS (/v2/speak PCM)
Endpointing (150ms)
Deepgram Nova-3 WS
Real-time (0ms)
Groq gpt-oss-20b
Deepgram Flux (/v2/speak)
24kHz PCM Audio
Waiting for user speech...
Waiting for response generation...
| Turn | VAD (ms) | STT (ms) | Groq TTFT | LLM Total | Flux TTFA | Total (ms) |
|---|---|---|---|---|---|---|
| No turns recorded yet. Click 'Start Speaking' to test the optimized pipeline with Deepgram Flux TTS. | ||||||