- NEW
- LIST
- DATA
- HUMAN
Short answer
How do I make my AI sound like a specific person?
You provide audio samples to your agent, which then uses the LMNT connector to build a unique voice profile. You can then list or retrieve these profiles to use them for speech synthesis. You get full control over your agent's vocal identity.
Voice profile outcomes
What you can do with custom voices
The AI handles the heavy lifting of voice management.
NEW
Custom voice profiles
The AI builds a unique vocal identity from your audio. You get a specific profile you can call whenever you need it.
LIST
Voice inventory
Your agent can see every voice profile you have already created. This makes it easy to switch personas mid-conversation.
DATA
Voice details
You can pull specific metadata for any existing profile. This ensures your agent uses the exact settings you intended.
HUMAN
Audio preparation
The AI cannot clean up your audio files. You need to provide clear, high-quality voice samples for the cloning to work.
The workflow
What your AI does when you provide audio.
Your agent uses the LMNT tools to manage your vocal assets.
Build the profile
The agent processes your audio samples to generate a brand new voice identity.
create_voiceCheck your library
The agent looks through your stored voice profiles to see what is available.
list_voicesVerify voice settings
The AI retrieves the specific configuration and ID for a single voice profile.
get_voiceApply the voice
The agent uses the identified voice profile to perform speech synthesis tasks.
get_voice
Try it
Copy these to start.
Use these prompts to trigger the LMNT tools.
Starting points
These are opening lines. Replace the bracketed text with your actual file paths or voice names.
LMNT Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_HUL7nYTyeZDkCPGI9P4KyMLRDb0Pv4CZVWjuWJeF/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 — LMNT capabilities are ready to use.
{
"mcpServers": {
"lmnt-ultra-low-latency-speech-synthesis-mcp": {
"url": "https://edge.vinkius.com/vk_preview_HUL7nYTyeZDkCPGI9P4KyMLRDb0Pv4CZVWjuWJeF/mcp"
}
}
}Create a new voice profile using the audio file at /users/downloads/sample.wav.
Show me a list of all the voice profiles I have created so far.
Get the details for the voice profile named 'Narrator_01'.
Find the ID for my 'Assistant_Voice' so I can use it for this next task.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect LMNT once, then ask.
Link your account once. Your credentials stay encrypted, and you can immediately start managing voices through your AI client.
Connect LMNT to your AIFAQ
How the voice cloning behaves
- 01
Can the AI fix noisy audio files?
No. You must provide clean audio samples for the voice creation process to work.
- 02
Can I delete voice profiles through the AI?
No. The current tools only allow you to create, list, and retrieve voice information.
- 03
How do I know which voices are ready to use?
You can ask your agent to list all available voices to see your current inventory.
- 04
Does the AI store my raw audio files?
No. The AI uses the files to create the profile, but it does not act as a storage drive for your raw audio.
- 05
Can I edit an existing voice profile?
No. You can only create new profiles or retrieve information about existing ones.
More questions about LMNT? The Connector page answers them. See everything the LMNT Connector can do
Connect LMNT to Claude, Cursor, ChatGPT & more
