by Deepgram
Deepgram offers speech-to-text, text-to-speech and Voice Agent APIs for real-time voice applications. The Voice Agent API handles listening, language model orchestration and spoken replies in one stack.
APIUsage-based
Step 1: Request a demo or sign up with Deepgram.
Step 2: Choose speech-to-text, text-to-speech or the Voice Agent API.
Step 3: Connect your language model and business logic.
Step 4: Stream audio in and play spoken replies as they arrive.
Step 5: Add function calling for actions like booking appointments.
It combines speech-to-text, LLM orchestration and text-to-speech in one platform.
The site shows sending each chatbot message to Deepgram TTS over WebSocket and playing it in the browser.
Yes. The example receptionist books appointments through function calling and confirms details before booking.
DeepgramDeepgram provides APIs for speech-to-text, text-to-speech and voice agents. It is aimed at developers and companies building real-time voice products.
Speechmatics Voice AISpeechmatics provides speech-to-text APIs for real conversations, handling accents, multiple speakers and multilingual speech. It runs on-device, on-prem or in the cloud.