# ElevenLabs TTS V3

> Audio AI Model by ElevenLabs — available on Layer

ElevenLabs' most expressive TTS model with inline audio tags for emotion and delivery.

Eleven v3 is ElevenLabs' most advanced text-to-speech model, built from the ground up for emotional expressiveness. Voices can sigh, whisper, laugh, and react using inline audio tags like `[whispers]`, `[laughs]`, and `[excited]`. Supporting 70+ languages with multi-speaker dialogue capabilities via the Text to Dialogue API, it is best suited for audiobooks, film, dramatic voiceovers, and any content requiring nuanced vocal performance.

## Specifications

| Property | Value |
|----------|-------|
| Provider | ElevenLabs |
| Category | Audio |
| Price | 0.03 / character count per character_count |

## Capabilities

- Text to Speech
- Auto Duration

---

[Try ElevenLabs TTS V3](https://next.app.layer.ai/me/session/new?baseModelId=ELEVEN_LABS_TTS_V3) | [All Audio Models](https://layer.ai/models/audio) | [All Models](https://layer.ai/models)