cartesia-mcp
The official Cartesia MCP Server
Documentation
Cartesia MCP Server
The Cartesia MCP server exposes Cartesia APIs over the Model Context Protocol (MCP) so clients such as Cursor, Claude Desktop, and OpenAI Agents can list voices, run TTS and STT, manage pronunciation dictionaries, clone voices, and more—without one-off scripts.
Documentation: Cartesia docs — MCP
Requirements
- **uv** — runs the server via `uvx` with no global install
- Python 3.13+ (installed automatically by `uvx`)
- A **Cartesia API key** for TTS, STT, voices, and related APIs
- Optionally, an **admin API key** (Keys → Admin) for management tools such as `get_credit_usage`. Admin keys and standard keys are separate credentials; each only works on its own route class.
Setup
Get an API key. Full instructions: Cartesia docs — MCP.
CLI (recommended) — `npx add-mcp "uvx cartesia-mcp" --name cartesia --env 'CARTESIA_API_KEY=${CARTESIA_API_KEY}'`
Cursor — Install Cartesia MCP, then set `CARTESIA_API_KEY` in Settings → MCP.
Claude Code — `claude mcp add -e CARTESIA_API_KEY= cartesia -- uvx cartesia-mcp`
Manual setup
Add to `.cursor/mcp.json`, `.mcp.json` (Claude Code), or your client’s MCP config:
{
"mcpServers": {
"cartesia": {
"command": "uvx",
"args": ["cartesia-mcp"],
"env": {
"CARTESIA_API_KEY": ""
}
}
}
}Try it
Ask your agent things like:
- List all available Cartesia voices
- Convert text to audio with a chosen voice (speed, volume, emotion)
- Transcribe an audio file to text
- Create a pronunciation dictionary and use it in TTS
- Check credit usage for your account
- Localize an existing voice into another language
- Change an audio file to use a different voice
Tools
| Tool | Description |
|---|---|
| `text_to_speech` | Convert text to audio; optional speed, volume, emotion, and pronunciation dict. Default `save=true` returns `file_id` and a 24h `download_url`. |
| `speech_to_text` | Transcribe an audio file (`mode=batch` default, or `mode=stream`) |
| `list_voices` | List available voices (filter by language, search, gender, etc.) |
| `get_voice` | Fetch metadata for a voice by ID |
| `clone_voice` | Clone a voice from an audio sample |
| `update_voice` | Update a cloned voice's name or description |
| `delete_voice` | Delete a cloned voice |
| `voice_change` | Re-render audio with a different voice |
| `localize_voice` | Adapt a voice to another language or dialect |
| `add_voice_accents` | Add catalog accents to an instant voice clone (`british`, `parisian`, …) |
| `delete_voice_accent` | Remove a catalog accent from an instant voice clone |
| `list_pronunciation_dicts` | List pronunciation dictionaries |
| `create_pronunciation_dict` | Create a pronunciation dictionary |
| `get_pronunciation_dict` | Get a pronunciation dictionary by ID |
| `update_pronunciation_dict` | Update a pronunciation dictionary |
| `delete_pronunciation_dict` | Delete a pronunciation dictionary |
| `download_file` | Fetch a cloud file by ID (`download_url` + local copy) |
| `get_credit_usage` | Credit usage over time (`CARTESIA_ADMIN_API_KEY`) |
See `cartesia_mcp/server.py` for parameters and return types.
Releases
Versions and PyPI publishes are driven by Conventional Commits on `main` via release-please. Use PR titles like `feat: …` or `fix: …` (especially when squash merging). See CONTRIBUTING.md.
Local development
Run your checkout in an MCP client instead of the published `uvx cartesia-mcp` package:
git clone https://github.com/cartesia-ai/cartesia-mcp.git
cd cartesia-mcp
uv sync --devSet `CARTESIA_API_KEY` (and optionally `CARTESIA_ADMIN_API_KEY`). Replace `/path/to/cartesia-mcp` below with your checkout path.
Cursor — add to `.cursor/mcp.json`:
{
"mcpServers": {
"cartesia": {
"command": "uv",
"args": ["--directory", "/path/to/cartesia-mcp", "run", "cartesia-mcp"],
"env": {
"CARTESIA_API_KEY": ""
}
}
}
}Restart Cursor or reload MCP servers, then confirm cartesia appears under Settings → MCP.
Claude Code:
claude mcp add -e CARTESIA_API_KEY= cartesia -- uv --directory /path/to/cartesia-mcp run cartesia-mcpIn a Claude Code session, run `/mcp`, select cartesia, and verify tools load.
Testing
Unit tests (no API keys):
uv sync --dev
uv run pytestSmoke-test all tools (requires `CARTESIA_API_KEY`):
uv run python scripts/test_all_tools.pyThe script creates temporary cloned/localized voices and pronunciation dictionaries, then deletes only those. It does not delete catalog or other existing resources.
Advanced
Output directory
By default, generated audio is written to the server's working directory. To choose a fixed folder, add `OUTPUT_DIRECTORY` to `env`:
"env": {
"CARTESIA_API_KEY": "",
"OUTPUT_DIRECTORY": "~/cartesia-output"
}Local audio files
Tools like `speech_to_text` and `voice_change` need paths to existing audio files on disk. Pass the full path to each file when prompting your agent. For `speech_to_text`, use the default batch mode for common containers (mp3, flac, wav, etc.). Use `mode="stream"` for mono PCM WAV or raw PCM with `encoding` and `sample_rate`.
Admin API key
Some tools call management endpoints that accept admin API keys only (`sk_car_admin_...`). Set `CARTESIA_ADMIN_API_KEY` in `env` alongside `CARTESIA_API_KEY`:
- `CARTESIA_API_KEY` — TTS, STT, voices, pronunciation dictionaries, voice changer, etc.
- `CARTESIA_ADMIN_API_KEY` — optional; required for `get_credit_usage` today. Admin keys do not work on generation routes, and standard keys do not work on admin routes.
Mint admin keys in the Playground under Keys → Admin (org admins only).
Hosted OAuth redirect URIs
Hosted MCP (`mcp.cartesia.ai`) accepts Dynamic Client Registration with a restricted redirect-URI policy:
- Custom schemes (desktop apps) — e.g. `cursor://…`, `vscode://…`
- Loopback HTTP — `http://localhost|127.0.0.1|::1` (any port/path)
- Allowlisted HTTPS — first-party callbacks for Claude, ChatGPT, Cursor web/Agents, and VS Code Web
To temporarily allow another exact HTTPS callback without a code change, set:
MCP_OAUTH_EXTRA_HTTPS_REDIRECTS=partner.example|/mcp/oauth/callback(comma-separated `host|/path` pairs). Prefer adding durable hosts in code for known products.
API version
All tools send `Cartesia-Version` (default `2026-08-14`, the latest in Cartesia docs). Override with `CARTESIA_VERSION` in `env` if you pin an older integration date.
Frequently asked questions
What is cartesia-mcp?
cartesia-mcp is The official Cartesia MCP Server
How do I install cartesia-mcp?
Open the GitHub repository and follow its README. Most MCP servers are added to your client's MCP config, then called by your agent.
Is cartesia-mcp open source?
Yes — it is hosted on GitHub at https://github.com/cartesia-ai/cartesia-mcp and has 8 stars.
Related MCP tools
Damn Vulnerable MCP Server Python-based implementation. Trusted by 1200+ developers. Trusted by 1200+ developers. Trusted by 1200+ developers.
A Model Context Protocol (MCP) server that enables secure interaction with MySQL databases Python-based implementation. Trusted by 900+ developers.
Query MCP enables end-to-end management of Supabase via chat interface: read & write query executions, management API support, automatic migration versioning...
Model Context Protocol with Neo4j Python-based implementation. Trusted by 700+ developers. Trusted by 700+ developers. Trusted by 700+ developers.
An MCP server that provides control over Android devices via adb Python-based implementation. Trusted by 500+ developers.
A Model Context Protocol (MCP) server for PostgreSQL databases with enhanced capabilities for AI agents. Python-based implementation.
Run your own MCP server? See who uses it and what to fix.
Measure it with TrackMCP