- NEW VOICE
- LIBRARY
- DETAILS
- HUMAN
Short answer
How can I clone voices with my AI?
You provide audio samples to your agent, which uses Cartesia to generate a custom voice persona. You can then manage these voices through your chat interface. This breakdown shows how the tools work together.
Where every voice lands.
Voice management outcomes
Your agent organizes everything it creates through Cartesia.
NEW VOICE
Custom personas created
The AI uses your audio samples to build a new voice. You get a unique identity for your agent to use in speech tasks.
LIBRARY
Voice list organized
Your agent keeps track of every persona it has made. You can see all available voices through a single request.
DETAILS
Voice data retrieved
The AI pulls specific settings for a single voice. This ensures the agent uses the exact parameters you intended.
HUMAN
Manual audio review
The AI cannot judge if a voice sounds natural to a human ear. You must listen to the samples to verify the quality.
The workflow
What your AI does when you provide audio.
The AI handles the technical heavy lifting of voice synthesis and management.
Generate persona
The agent processes your audio clips to build a new voice profile.
clone_voiceCheck library
Your agent looks through your existing collection to see what is available.
list_voicesFetch settings
The agent pulls the specific configuration for a chosen voice.
get_voiceVerify identity
The agent confirms the voice exists before attempting to use it for speech.
get_voice
Try it
Copy these to start.
Use these prompts to trigger the Cartesia tools.
Starting points
These are starting points. You should replace the bracketed text with your actual audio file names or voice IDs.
Cartesia Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_fqUkC4o6TEWq9kAU8DlY8MAb7pfJxo3R952S85p7/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — Cartesia capabilities are ready to use.
{
"mcpServers": {
"cartesia-voice-ai-mcp": {
"url": "https://edge.vinkius.com/vk_preview_fqUkC4o6TEWq9kAU8DlY8MAb7pfJxo3R952S85p7/mcp"
}
}
}Clone a new voice using the audio file 'interview_sample.wav'.
Show me a list of all the voices I have created so far.
Get the details for the voice named 'narrator_v1'.
List my current voices and tell me which one sounds most professional.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect Cartesia once, then ask.
Just link your account once. Your credentials stay encrypted, and you can immediately start managing voices through your preferred AI client.
Connect Cartesia to your AIFAQ
How the voice cloning works
- 01
Can the AI delete a voice if I ask it to?
No. There is no tool provided to delete or remove voices from your library.
- 02
How long does it take to clone a voice?
The AI processes the audio samples provided, but the exact timing depends on the length of your samples and the agent's processing speed.
- 03
Can I edit the pitch of a voice after it is cloned?
No. The tools only allow for creating, listing, and retrieving voice data.
- 04
Do I need to upload audio every time I want to speak?
No. Once a voice is cloned, you can retrieve it by name or ID using the existing tools.
- 05
Can the AI create a voice from a text description alone?
No. You must provide audio samples to use the cloning capability.
More questions about Cartesia? The Connector page answers them. See everything the Cartesia Connector can do
Connect Cartesia to Claude, Cursor, ChatGPT & more
