← All posts
· 4 min read

Use LeanVox TTS Directly from Claude, ChatGPT & Cursor — No Code Required

LeanVox now has native MCP support. Add one config block and your AI assistant can generate speech, create podcasts, and clone voices — zero code.

mcpaittsclaudedeveloper-tools

What if your AI assistant could just… talk?

Not "here's a link to an audio file." Actually generate speech, mid-conversation, without you writing a single line of code.

That's what our MCP server does.

npx leanvox-mcp

Add it to Claude Desktop, Cursor, VS Code, or any MCP-compatible client. Your AI can now generate speech, create podcasts, clone voices, and check your balance — all through natural conversation.

What is MCP?

Model Context Protocol is an open standard (created by Anthropic) that lets AI apps connect to external tools. Think of it as USB-C for AI — one plug, works everywhere.

Claude Desktop, ChatGPT, Cursor, Windsurf, VS Code, and hundreds of other apps support it. One MCP server = instant integration with all of them.

Setup: 30 Seconds

Claude Desktop — edit claude_desktop_config.json:

{
  "mcpServers": {
    "leanvox": {
      "command": "npx",
      "args": ["leanvox-mcp"],
      "env": {
        "LEANVOX_API_KEY": "lv_live_your_key"
      }
    }
  }
}

Cursor — add to MCP settings. VS Code — add to Copilot MCP config. Same JSON, same result.

Restart. Done. Your AI can now speak.

What Your AI Can Do

5 tools available:

ToolWhat it does
leanvox_generateGenerate speech from text
leanvox_streamStream audio to a file
leanvox_dialogueCreate multi-speaker conversations
leanvox_list_voicesBrowse available voices
leanvox_check_balanceCheck your credit balance

3 built-in prompts:

  • "narrate" — Convert any text to natural speech
  • "podcast" — Turn content into a multi-speaker podcast
  • "voice-clone" — Clone a voice from a reference file

Just ask Claude: "Read this article aloud in a warm female voice" or "Turn this changelog into a podcast episode." It handles everything.

Why MCP > Writing Code

Most TTS integrations require: get API key, install SDK, write code, handle audio output, debug, iterate, deploy.

With MCP: add config, ask your AI. That's it. Zero code. The AI handles the API calls, parameter selection, and audio delivery.

What It Costs

Same pricing as the API — pay-per-character, no subscription:

  • Standard: $0.005/1K characters
  • Pro (with cloning): $0.01/1K characters
  • Credits never expire
  • $0.50 free on signup

A 500-word article narration costs ~$0.015. A 5-minute podcast episode costs ~$0.05.

The Bigger Picture

We think TTS should be a native capability of AI agents, not an afterthought. MCP makes that possible.

LeanVox is the first TTS provider with native MCP support. No wrappers, no middleware — a purpose-built MCP server that exposes the full API.

Get your API key · GitHub · npm


Try LeanVox free

$0.50 in free credits. No credit card required. Start generating speech in 30 seconds.

Get started free →