Introducing Solaria-2. Our newest English speech-to-text model purpose-built for Voice AI. Learn more here
WebSocket to send audio chunks and receive events.