Review · Software & AI
Deepgram Voice Agent API review
A single WebSocket API that runs speech-to-text, LLM orchestration and text-to-speech, billed per minute of connection time.
By the DBotMaker research desk · Updated November 7, 2026
- Capabilities (40%)
- 7.9
- Ease of setup (30%)
- 7.3
- Value (30%)
- 8.9
Deepgram's Voice Agent API is infrastructure rather than a dashboard. One WebSocket takes audio in and returns audio out, while Deepgram runs speech-to-text, LLM orchestration, text-to-speech, barge-in and function calling. Pricing is per minute of connection time: $0.075 per minute covers its STT, a managed LLM and TTS, bringing your own LLM or voice drops it to as low as $0.050, and an Advanced tier costs $0.163. New accounts get $200 in free credit with no expiry. Flux turn detection ends a turn in about 260 ms, every turn returns a latency breakdown, and automatic fallback keeps a model failure from ending the call. You can self-host on Kubernetes or pick Global, EU, India or Australia endpoints, and agents understand 60+ languages. Deepgram says over 1,000 organizations have built on the API. The catch for agencies is that there are no numbers, campaigns or visual builder, so you supply telephony and the app logic.
Who it is for
Engineering teams who want a voice agent API they can self-host or bring their own LLM and voice to.
Price
Pay as you go with $200 in free credit; $0.075 per minute standard, as low as $0.050 per minute with your own LLM or voice.
Bring your own LLM or voice and it drops to as low as $0.050/min.
Quoted from the vendor pricing page on November 7, 2026.
Key features
- One WebSocket for STT, LLM orchestration and TTS
- Flux turn detection with end of turn in about 260 ms
- Automatic LLM and TTS fallback mid-call
- Self-hosting on Kubernetes with a Helm chart
- Global, EU, India and Australia endpoints
Strengths and weaknesses
- Strength: $200 free credit that does not expire
- Strength: Per-turn latency report on by default
- Strength: Self-hosting option for data control
- Weakness: No phone numbers or call-flow builder; you wire up telephony
- Weakness: Advanced model tier costs $0.163 per minute
- Weakness: BAAs are for Enterprise customers
Key facts
- Free credit: $200
- Languages understood: 60+
- Organizations building: 1,000+, per the vendor
Where it ranks
Deepgram Voice Agent API is No. 8 in our Best AI Voice Agent Platforms verdict and No. 6 in Best AI Voice Agents.
Alternatives we scored
- Retell AI8.9
- ElevenLabs Agents8.8
- Vapi8.8
- Telnyx Voice AI8.7
Visit deepgram.com · Questions or corrections: editorial@dbotmaker.io