What is Deepgram Voice AI?

Deepgram Voice AI is a suite of real‑time and batch APIs that provide speech‑to‑text, text‑to‑speech, and voice agent capabilities. The Speech‑to‑Text API detects natural speech, identifies turn boundaries, and streams transcripts with low latency. The Text‑to‑Speech API converts generated text into natural‑sounding audio in multiple voices and languages.

The Voice Agent API unifies STT, TTS, and LLM orchestration, managing conversation history, context, function calls, and external system integration. The platform supports cloud, on‑premise, and telephony (PSTN/SIP) deployments, enabling contact centers, customer support, medical transcription, and podcast workflows.

Deepgram Voice AI pricing Freemium

Pay as you go $0

Growth $4k+/year

Custom - byo llm + tts $0.050/min

Custom - byo llm $0.056/min

Nova-1 & 2 $0.0058/min

Byo tts $0.065/min

Voice agent api $0.075/min

Standard $0.075/min

Streaming $0.0077/min

Flux $0.0077/min

Nova-3 (monolingual) $0.0077/min

Custom model $0.0077/min

Nova-3 (multilingual) $0.0092/min

Base $0.0145/min

Advanced $0.163/min

Enhanced $0.0165/min

Text-to-speech $0.030/1k characters

Aura-2 $0.030/1k characters

Aura-1 $0.0150/1k characters

Enterprise Contact us

Verify on the official pricing page.

View plans

Deepgram Voice AI user reviews

Would you recommend Deepgram Voice AI?

Recommend this tool?

Deepgram Voice AI's key features

Speech-to-text API
Text-to-speech API
AI voice agent API
Multi-tenant cloud deployment
Self-hosted deployment options
40x faster inference
30% higher accuracy

Deepgram Voice AI use cases

Real‑time speech transcription for contact centers to provide instant, low‑latency transcripts and sentiment insights for agents, improving customer satisfaction without additional tools.
Batch speech‑to‑text API to convert entire podcast libraries into searchable, SEO‑optimized transcripts, enabling quick content discovery and multi‑language subtitles.
Secure, HIPAA‑compliant medical transcription and text‑to‑speech synthesis, turning doctor‑patient conversations into accurate records and voice‑guided patient instructions while ensuring privacy.