> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.resemble.ai/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.resemble.ai/_mcp/server.

# Model Versions

**Resemble Ultra** is the current text-to-speech model for new and upgraded voices. The synthesis API automatically uses the model associated with your `voice_uuid`, so you do not need to select a model in the request.

`resemble-ultra` identifies the voice model version. Do not pass it as a `model` field when making text-to-speech requests.

## Resemble Ultra

| Version Name   | Version Code     | Status              | Streaming Support |
| -------------- | ---------------- | ------------------- | ----------------- |
| Resemble Ultra | `resemble-ultra` | Current and default | Yes               |

Resemble Ultra is available through synchronous synthesis, HTTP streaming, and WebSocket streaming. To move an existing voice to Ultra, visit the [Voices page](https://app.resemble.ai/hub/voices) and upgrade the voice.

## End-of-Life Text-to-Speech Models

All previous Resemble text-to-speech models have reached end of life. Voices that still use one of these versions can no longer generate audio and must be upgraded to Resemble Ultra.

The historical version codes below are listed for reference only. They are not valid model selections for synthesis requests.

| Version Name             | Historical Version Code | Status      |
| ------------------------ | ----------------------- | ----------- |
| Chatterbox               | `tts-v4`                | End of life |
| Chatterbox-Turbo         | `tts-v4-turbo`          | End of life |
| Chatterbox Multilingual  | `tts-v4`                | End of life |
| Resemble Enhanced TTS v3 | `tts-v3`                | End of life |
| Resemble Enhanced TTS v2 | `tts-v2`                | End of life |
| Resemble Enhanced TTS v1 | `tts-v1`                | End of life |
| Resemble Legacy TTS      | `tts-legacy`            | End of life |

## Speech-to-Speech Models

| Version Name         | Version Code | Description                                                | Dataset Requirements | Streaming Support | Release Date |
| -------------------- | ------------ | ---------------------------------------------------------- | -------------------- | ----------------- | ------------ |
| Resemble Core STS v2 | `sts-v2`     | Improved pitch tracking and 48 kHz support.                | 10+ minutes          | ✅ Yes             | Q4 2023      |
| Resemble Core STS v1 | `sts-v1`     | Enhanced conversion with better speed and accuracy.        | 10+ minutes          | ✅ Yes             | Q2 2023      |
| Resemble Legacy STS  | `sts-legacy` | Initial speech-to-speech model for basic voice conversion. | 10+ minutes          | ✅ Yes             | Q2 2021      |