- AUDIO
- VOICE
- DATA
- HUMAN
Short answer
How can I turn text into speech using my AI?
Your agent uses LMNT to convert text into audio streams by calling specific voice IDs. You can pick from a list of available voices to get the sound you want. This breakdown shows how the tools work together.
The outcomes.
What your request produces.
Depending on your prompt, your agent delivers one of these results.
AUDIO
Spoken audio streams
The AI generates audio from your text. You get a playable stream based on the voice you selected.
VOICE
Voice selection
The AI identifies which voices are available. You can choose a specific ID to match the tone you want.
DATA
Voice details
The AI pulls specific attributes for a voice. This helps you understand the characteristics of a voice before you use it.
HUMAN
Manual voice selection
The AI cannot decide which voice sounds best for your specific context. You must tell the agent which voice ID to use.
The workflow
What your AI does when the text arrives.
The AI manages the technical calls to LMNT to handle the conversion.
Find a voice
The AI looks through the available library to see which voices are ready to use.
list_voicesCheck voice details
If you need more info, the AI pulls the specific properties of a single voice ID.
get_voiceGenerate audio
The AI sends your text and the chosen voice ID to create the audio stream.
generate_speechDeliver stream
The AI provides the resulting audio data back to your chat interface.
generate_speech
Try it
Copy these to start.
Paste these directly into your AI client to begin.
Starting points
These are opening lines. Swap out the text or voice names for your own.
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"
}
}
}List all available voices in LMNT.
Use the voice with ID 'en-US-standard-A' to read this paragraph: 'The quick brown fox jumps over the lazy dog.'
Get the details for the voice ID 'premium-male-01'.
Read the contents of my 'meeting_notes.txt' file using a female voice from the list.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect LMNT once, then ask.
You only need to link the MCP once. Your credentials stay encrypted, and you can immediately start asking your agent to speak.
Connect LMNT to your AIFAQ
How this task behaves.
- 01
Can the AI choose the voice by itself?
No. The AI can list the voices, but you need to tell it which specific voice ID to use for the speech generation.
- 02
Can the AI edit the audio file after it is made?
No. The tools only handle the generation of the speech from text. You would need to use other tools for audio editing.
- 03
How do I know which voices are available?
You can ask your AI to list the voices, and it will use the list tool to show you the options.
- 04
Can the AI change the pitch or speed of the voice?
No. The current tools only support generating speech using the provided voice IDs.
- 05
Does the AI save the audio files for me?
No. The AI generates the audio stream based on your request, but it does not manage file storage.
More questions about LMNT? The Connector page answers them. See everything the LMNT Connector can do
Connect LMNT to Claude, Cursor, ChatGPT & more
