elevenlabs.io llms.txt

llms.txt 11 KB, 114 lines

# ElevenLabs

ElevenLabs is an AI research and product company transforming how we interact with technology. We launched in January 2023 with the first human-like AI voice model. Today, we serve millions of users and thousands of businesses across three main platforms. ElevenAgents enables businesses to deliver […] and intelligent customer experiences, with the integrations, testing, monitoring, and reliability necessary to deploy voice and chat agents at scale. ElevenCreative […] creators and marketers to generate and edit speech, music, image, and video across 70+ languages. ElevenAPI gives developers access to our leading AI audio foundational models.

> [!NOTE]
> ## ElevenLabs MCP server […] [elevenlabs.io/mcp](https://elevenlabs.io/mcp)
>
> Hosted, remote MCP server over HTTP. Exposes ElevenAgents management and ElevenCreative generation (50+ models) as tools to any MCP client, scoped to the authenticated workspace.
>
> **Endpoints**
> - `https://api.elevenlabs.io/v1/mcp` […] global
> - `https://api.us.elevenlabs.io/v1/mcp` […] US residency
>
> **Auth:** OAuth, one time per client. No API keys to store or rotate; access is revocable from the workspace.
>
> **Capabilities**
> - *Agents:* list and query agents, read conversation transcripts and call metrics (latency, outcomes, failures), create agents, update prompts, voices, and tool configuration. Changes take effect without a redeploy.
> - *Creative:* generate speech, music, images, and video. Regenerate an individual layer (voiceover, music, single shot) without re-running the rest.
>
> **Setup:** per-client install steps (ChatGPT, Claude, Claude Code, Cursor, GrokBot, Hermes) and the ElevenLabs CLI are documented at [elevenlabs.io/mcp](https://elevenlabs.io/mcp). Fetch it with `Accept: text/markdown` for the current commands and config.

## Fetching Markdown Content

Most ElevenLabs pages support a Markdown-optimized response format. When you request a page with the header `Accept: text/markdown`, the server returns a stripped-down Markdown version of the page […] no navigation or boilerplate […] containing only the essential content. This significantly reduces token usage and improves parsing accuracy.

**To request Markdown content, add this header to your HTTP requests:**

```
Accept: text/markdown
```

**Example (curl):**
```bash
curl -H "Accept: text/markdown" https://elevenlabs.io/text-to-speech
```


Where Markdown responses are available, the server will reply with `Content-Type: text/markdown`. If a page does not support this format, it will fall back to the standard HTML response. The [LLM-optimized full documentation](https://elevenlabs.io/docs/llms-full.txt) is always available as a single flat Markdown file without requiring any special headers.

## Products

### ElevenCreative
- [ElevenCreative overview](https://elevenlabs.io/creative): A single studio for creators to combine voice, music, image, and video. Built for creators, brands, and media companies […] storytelling and production workflows.
- [Text to Speech](https://elevenlabs.io/text-to-speech): Controllable, expressive speech across 70+ languages. Our models dynamically predict voice characteristics and adapt to content context.
- [Speech to Text](https://elevenlabs.io/speech-to-text): Scribe, our most accurate Speech to Text model, with speaker diarization and character-level timestamps.
- [Voice Library](https://elevenlabs.io/voice-library): Thousands of pre-built voices, plus tools to clone your own voice or design one from a prompt.
- [Voice Cloning](https://elevenlabs.io/voice-cloning): Create a high-fidelity replica of a voice from audio samples. Requires consent verification for cloning public figures.
- [Music](https://elevenlabs.io/music): Studio-quality music from natural language prompts in any genre or style, trained on licensed data and cleared for commercial use.
- [Sound Effects](https://elevenlabs.io/sound-effects): Custom sound effects and ambient audio from text prompts.
- [Productions](https://elevenlabs.io/productions): Translate and re-voice video and audio content across languages while preserving original speaker characteristics.
- [Projects / Studio](https://elevenlabs.io/studio): Create and edit audiobooks, podcasts, and voiceovers in an editor built on all of ElevenLabs' audio research.
- [Video](https://elevenlabs.io/video): Turn ideas into videos using leading models including Veo, Sora, Wan, Kling, and Seedance.
- [Image](https://elevenlabs.io/image): Generate and edit images with leading AI image models.

### ElevenAgents
- [ElevenAgents overview](https://elevenlabs.io/agents): Enterprise platform for AI agents that can talk, type, and take action. […] customer experience, […] revenue teams, and […] internal operations.
- [Conversational AI](https://elevenlabs.io/conversational-ai): Build agents powered by V3, our most expressive text to speech model, with improved turn-taking, natural dialogue, and emotion understanding.
- [Agents Platform](https://elevenlabs.io/docs/eleven-agents/overview): Configure multimodal agents via dashboard, developer toolkit, or visual workflow builder. Deploy across telephony, web, and mobile.
- [LLM Integration](https://elevenlabs.io/docs/eleven-agents/customization/llm): Agents support natively integrated LLMs (OpenAI, Claude, Gemini, and others) plus custom LLM endpoints. Backup LLM cascading is configurable to maintain conversation continuity if a primary model fails.

First 59 of 114 lines. 13 spans shown as […] because the words or characters in them are rejected by this site's own style rules; the live file has them.