The Cartesia MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Cartesia account. It provides structured and secure access to voice, speech synthesis, pronunciation dictionary, accent, and managed agent model capabilities, so your agent can generate audio from text, find suitable voices, check supported accents, manage pronunciation dictionaries, and review available agent models on your behalf.
- Text-to-speech generation: Have your agent turn a complete transcript into a downloadable audio file using a selected Cartesia voice.
- Voice discovery: Let the agent search Cartesia voices and retrieve synthesis details to help choose the right voice for an audio project.
- Accent exploration: Direct your agent to list supported accents and filter them by language, locale, or voice localization support.
- Pronunciation dictionary management: Instruct your agent to create, review, update, list, or delete reusable pronunciation dictionaries for more consistent speech generation.
- Managed agent model selection: Have your agent review available Cartesia managed agent models, including provider, latency, and pricing details.