> ## Documentation Index
> Fetch the complete documentation index at: https://heygen-1fa696a7.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# HeyGen Voice

> HeyGen Voice is HeyGen's in-house voice model: professional voice clones trained per speaker, with completed or streaming speech synthesis.

<img className="w-full h-44 object-cover rounded-xl" src="https://mintcdn.com/heygen-1fa696a7/hfMXXwJzjE7vBSYZ/images/theme/research-3.webp?fit=max&auto=format&n=hfMXXwJzjE7vBSYZ&q=85&s=c8191ba4743fd85050f1af1a93ed4969" alt="" noZoom width="1400" height="788" data-path="images/theme/research-3.webp" />

<Warning>
  HeyGen Voice is in private preview and is a **paid feature**. Your account must be enabled ([request access](/docs/voices/professional-voice-clone#get-private-preview-access)), and each professional voice occupies a purchased voice clone slot. Buy slots on the [API usage page](https://app.heygen.com/developers/usage).
</Warning>

HeyGen Voice trains a dedicated voice adapter per speaker and synthesizes speech directly from it. It is the highest-fidelity voice output HeyGen offers.

* **Professional voice cloning** — train a voice from 1–10 recordings of the same speaker totaling at least 20 minutes, then poll until it is `ACTIVE`. See [HeyGen Professional Clone](/docs/voices/professional-voice-clone).
* **Speech generation** — synthesize from an `ACTIVE` voice as one 44.1 kHz WAV, or stream ordered audio parts over Server-Sent Events with optional word timestamps. See [HeyGen Voice Speech](/docs/voices/heygen-voice-speech).

To clone from a single short recording in minutes, use [HeyGen Instant Clone](/docs/voices/instant-voice-clone) instead. It runs on the Starfish engine, and its voices work everywhere a catalog `voice_id` does.

## Pricing

|            |                                                                                                                                 |
| ---------- | ------------------------------------------------------------------------------------------------------------------------------- |
| Voice slot | One purchased professional voice clone slot per voice, bought on the [API usage page](https://app.heygen.com/developers/usage). |
| Training   | Five pooled trainings per slot per monthly billing period, initial training included. Failed trainings are free.                |
| Synthesis  | 0.6 API credits per generated minute.                                                                                           |

A voice beyond the workspace's slot limit, for example after a slot add-on lapses, returns `voice_expired` on synthesis. Add a slot or delete another professional voice and it works again immediately.

## Endpoints

| Endpoint                                    | Purpose                                   |
| ------------------------------------------- | ----------------------------------------- |
| `POST /v3/models/audio/voices`              | Create or retrain a professional voice.   |
| `GET /v3/models/audio/voices/{voice_id}`    | Read a voice and its training status.     |
| `GET /v3/models/audio/voices`               | List your professional voices.            |
| `DELETE /v3/models/audio/voices/{voice_id}` | Delete a voice.                           |
| `POST /v3/models/audio/tts`                 | Generate one completed audio file.        |
| `POST /v3/models/audio/tts/stream`          | Stream audio parts as they are generated. |

Voices are scoped to the workspace that trained them. The [voice catalog](/docs/voices/overview) (`/v3/voices`) is the other surface: 300+ stock voices, [designed voices](/docs/voices/design-voices), and instant clones, all synthesized through [Third Party Speech](/docs/voices/speech).