# Sequencer

> Build real-time AI video avatars, voice-only chat and text conversations with the Sequencer API. Anima 2 turns portraits and speech into live avatar video.

## Live avatars

- [Live avatar API documentation](https://sequencer.media/live-avatar-docs.md): REST-first guide with authentication, avatar creation, voices, preparation, video, voice-only and text clients, external speech, history, billing and every workspace avatar endpoint. Includes copyable JavaScript, Python and cURL examples.
- [Live avatar docs (HTML)](https://sequencer.media/live-avatar-docs): Navigable documentation and searchable endpoint reference.
- [Live avatars](https://sequencer.media/live-avatar): Product overview.

## Developer resources

- [API documentation](https://sequencer.media/docs?tab=api): Complete API configuration reference. Base URL: https://api.sequencer.media/v1. Send x-api-key from your server.
- [Worldstream API](https://sequencer.media/worldstream-api): Interactive world integrations.
- [MCP setup](https://sequencer.media/mcp): Optional assistant-driven tools. Remote endpoint: https://mcp.sequencer.media/.

## Integration workflow

Read the live avatar Markdown guide. List workspaces and preset voices, create or reuse an avatar, then choose video, voice-only or text channels. Prepare video assets before video calls. Keep API keys on your server; return user-scoped session credentials to the browser. Video uses LiveKit. Voice and text use conversationStream over WebSocket. End sessions after use and on failure. Use MCP when assistant-driven setup is requested.