# Voice Agent APIs

> Source: https://theagentsindex.com/collections/voice-agent-apis (curated, quality-gated, re-verified)

This collection rounds up the voice agent platforms in our index that expose a programmatic integration surface to developers. Each tool here ships with a real interface — a REST endpoint, a WebSocket, an SDK, an MCP server, or an equivalent — so you can wire its voice layer into your own product rather than only use the vendor's hosted console. Voice-only consumer products, no-API receptionists, and platforms that require a sales call before publishing any integration surface are excluded. The same voice model, behind a real developer API, qualifies; a great voice that ships without one does not.

| Name | What it is | Why it's here | Pricing | API |
| --- | --- | --- | --- | --- |
| [Bland AI](https://theagentsindex.com/bland-ai.md) | Enterprise voice AI that owns its own model stack, priced all-in per minute with no model pass-throughs. | Enterprise voice AI that owns its own model stack, priced all-in per minute with no model pass-throughs, and ships a public API for embedding per-minute voice agents into a third-party product. | — | Yes |
| [Deepgram Voice Agent API](https://theagentsindex.com/deepgram-voice-agent-api.md) | The speech-infrastructure vendor's own voice-agent API. Managed STT, LLM and TTS over a single WebSocket, $4.50/hr flat, with 6 LLM providers and self-hosted or VPC deployment on the same product. | The speech-infrastructure vendor's own voice-agent product: a single WebSocket drives managed STT, LLM and TTS, with six named LLM providers and a flat $4.50/hr rate. | — | Yes |
| [ElevenLabs Conversational AI](https://theagentsindex.com/elevenlabs-conversational-ai.md) | The voice-quality specialist’s own agent layer, 10,000+ voices and instant cloning, with a bring-your-own-LLM brain. | The voice-quality specialist's own agent layer, with a developer API exposing 10,000+ voices, instant cloning, and a bring-your-own-LLM brain behind it. | — | Yes |
| [OpenAI Realtime API](https://theagentsindex.com/openai-realtime-api.md) | OpenAI's own speech-to-speech model API for voice agents: WebRTC, WebSocket or native SIP telephony, billed purely on audio tokens with no platform fee. | OpenAI's own speech-to-speech model API for voice agents, reachable over WebRTC, WebSocket, or native SIP telephony, billed purely on audio tokens with no platform fee. | — | Yes |
| [PolyAI](https://theagentsindex.com/polyai.md) | The pre-LLM-era enterprise voice AI incumbent ($750M-valued, 200+ enterprise customers): sales-led only again after a May 2026 self-serve tier went away. | The pre-LLM-era enterprise voice AI incumbent ($750M-valued, 200+ enterprise customers), with a published developer surface for embedding its voice agents into a third-party product. | — | Yes |
| [Retell AI](https://theagentsindex.com/retell-ai.md) | Build phone-call voice agents on Retell's turn-taking layer, plug in your own LLM, voice, and telephony, ship over voice, chat, or SMS. | Phone-call voice agents built on Retell's turn-taking layer, with a public API to plug in your own LLM, voice, and telephony, shippable over voice, chat or SMS. | — | Yes |
| [Synthflow AI](https://theagentsindex.com/synthflow-ai.md) | A no-code voice AI platform for enterprise phone agents — a visual Flow Designer and a structured Build-Evaluate-Launch-Learn deployment cycle, billed by the minute in separate components. | A no-code voice AI platform for enterprise phone agents, with a visual Flow Designer and a developer API exposed alongside the hosted build surface. | — | Yes |
| [Cartesia](https://theagentsindex.com/cartesia.md) | The Stanford SSM-research spinout behind Sonic (TTS) and Ink (STT) — Line, its open-source voice-agent SDK, connects to 100+ LLMs and starts at a genuine $0/month, but the agent layer itself isn't self-hostable. | Line, Cartesia's voice-agent SDK, is Apache-2.0 with no caveat; ships as both an open SDK and a published API, the only Apache-licensed open-source voice-agent entry that does so. | — | Yes |
| [LiveKit Agents](https://theagentsindex.com/livekit-agents.md) | The Apache 2.0 voice-agent framework. 13.1k GitHub stars, self-hostable, and the only open-source pick in this category with fully published pricing. | Apache 2.0 voice-agent framework with 13.1k GitHub stars, self-hostable, with a documented API surface and the only open-source voice-agent pick with fully published pricing. | — | Yes |

## How this list is built

A tool earns a place here when its own Published, sourced listing names a programmatic integration surface a developer can call: a REST or GraphQL endpoint, a WebSocket, an SDK in at least one general-purpose language, an MCP server, or an equivalent open protocol. The API must be documented and reachable without a sales call — a "contact us" enterprise tier with no self-serve surface named anywhere in the listing does not qualify on its own. Voice-only consumer products that ship without a developer API are out, even where the underlying model is identical. Self-hosted open-source voice frameworks and hosted platforms are both included when each ships an API a developer can call; pricing model, rate limits and self-serve vs. sales-led onboarding vary entry by entry. Membership is re-checked against each listing's own already-researched, gate-passed fields rather than the `voice-agents` and `api` tags in isolation.