Review · AI tools
Deepgram Voice Agent API review
One WebSocket API that runs speech-to-text, LLM orchestration and text-to-speech for voice agents, with a self-hosted option.
By the TopToolAdvisor research desk · Updated November 12, 2026
- Features (35%)
- 8.4
- Ease of use (35%)
- 7.8
- Value (30%)
- 8.8
Deepgram's Voice Agent API is the most infrastructure-like option on this list, and it suits engineering teams that want to own the application and leave the speech loop to a vendor. One WebSocket takes audio in and returns audio out, while Deepgram runs end-of-turn detection, barge-in, function calling and conversation state. You can bring your own LLM through an OpenAI-compatible endpoint or Bedrock with ordered fallback, use a third-party voice, and change the model or voice mid-conversation. Every turn returns a latency report split by speech-to-text, LLM, tool and text-to-speech time. Pricing is per minute of connection: $0.075 on pay as you go with a managed LLM included, as low as $0.050 when you bring your own model or voice, and $0.163 for the advanced model tier, with $200 in free credit to start. Teams with strict data rules can self-host on Kubernetes with a Helm chart. You still need your own call handling and phone numbers.
Who it is for
Engineering teams that want an API, their own LLM and a self-hosting option
Price
$200 free credit; pay as you go from $0.075 per minute including speech and a managed LLM; from $0.050 per minute with your own LLM or voice.
Pay-as-you-go starts at $0.075/min, and that rate covers speech-to-text, a managed language model, and text-to-speech, with no separate provider bills.
Quoted from the vendor pricing page on November 12, 2026.
Key features
- Single WebSocket for speech-to-text, LLM orchestration and text-to-speech
- End-of-turn detection, barge-in and function calling handled by the API
- Bring your own LLM through an OpenAI-compatible endpoint or Bedrock, with ordered fallback
- Self-hosting on your own Kubernetes cluster with Deepgram's Helm chart
- Per-turn latency report broken down by speech, LLM, tool and voice time
Strengths and weaknesses
- Strength: $200 free credit with no credit card required
- Strength: Rate drops to as low as $0.050 per minute with your own LLM or voice
- Strength: Global, EU, India and Australia endpoints
- Weakness: API-first product with no visual agent builder mentioned on the product page
- Weakness: No phone numbers or telephony rates are listed with the Voice Agent API price
- Weakness: BAAs for HIPAA are signed for Enterprise customers only
Key facts
- Standard rate: $0.075 per minute, pay as you go
- Advanced model tier: $0.163 per minute
- Free credit: $200, no credit card
- Organizations building on it: 1,000+, per Deepgram
Where it ranks
Deepgram Voice Agent API is No. 5 in our Best AI Voice Agent Platforms verdict.
Alternatives we scored
- Retell AI8.8
- Vapi8.8
- ElevenLabs Agents8.7
- Telnyx Voice AI8.6
Visit deepgram.com · Questions or corrections: editorial@toptooladvisor.com