Kokoro

Kokoro on xpay — Fast, lightweight text-to-speech without the signup friction

Kokoro is a high-quality, lightweight text-to-speech engine built for speed and naturalness. Developers reach for it when they need to convert text into spoken audio without the overhead of larger TTS systems or the complexity of managing multiple API credentials. Kokoro powers applications ranging from accessibility features to voice-driven interfaces, and it's particularly popular for real-time or latency-sensitive use cases.

On xpay, you get Kokoro's capabilities as MCP-callable tools—no Kokoro account signup, no subscription, just pay $0.02 per synthesis call. Connect once with your xpay API key and start generating speech immediately.

What's available

• kokoro_tts ($0.02/call) — Kokoro 82M model for text-to-speech synthesis. Accepts text input and voice parameters, returns audio in your chosen format. Fast inference, natural prosody, multi-voice support.

When Kokoro is the right tool

• Real-time voice generation — Latency-sensitive applications where you need speech output in milliseconds, not seconds.
• Accessibility features — Adding audio narration to documents, articles, or web content for users who prefer listening.
• Voice-driven chatbots — Pairing LLM responses with natural speech for conversational AI agents.
• Lightweight deployments — When you want TTS without the resource footprint of larger models.
• Multi-voice applications — Generating different voices for different characters or personas in a single session.
• Cost-conscious projects — Per-call pricing means you only pay for what you use; no minimum spend or idle charges.

When to choose something else

• Enterprise voice cloning — If you need to clone a specific speaker's voice with high fidelity, specialized voice cloning APIs may be better suited.
• Extreme audio quality — For mastering-grade audio production, dedicated audio workstations or premium TTS services may offer more control.
• Offline-first requirements — If your application cannot make external API calls, a self-hosted TTS engine is necessary.
• Bulk batch processing — For processing millions of utterances, a batch API or downloadable model might be more economical than per-call pricing.

Pricing snapshot

Quickstart

Step 1: Get your xpay API key

Sign up at https://xpay.tools with email or Google. Copy your API key from the dashboard.

Step 2: Add Kokoro to Claude Code

claude mcp add --transport http kokoro "https://kokoro.mcp.xpay.sh/mcp?key=YOUR_XPAY_KEY"

Replace YOUR_XPAY_KEY with your actual key.

Step 3: Use it

Once connected, Claude Code will see the kokoro_tts tool. You can ask Claude to "generate speech from this text" or "create an audio file saying..." and it will invoke the tool automatically, billing you $0.02 per call.

For Cursor, Windsurf, Cline, or VS Code (Copilot):

Add this entry to your mcp.json file:

{
  "mcpServers": {
    "kokoro": {
      "url": "https://kokoro.mcp.xpay.sh/mcp?key=YOUR_XPAY_KEY"
    }
  }
}

Why route Kokoro through xpay?

• No Kokoro account needed — xpay holds the upstream credentials. You authenticate once with your xpay key and get instant access to Kokoro's API.
• Per-call pricing, no subscription — Pay $0.02 per synthesis. No monthly fees, no minimum spend, no overage surprises.
• MCP-native integration — Kokoro tools appear as native MCP resources in Claude Code, Cursor, and other MCP-compatible editors. No custom wrappers or API client code required.
• Unified billing — If you use multiple providers (Tavily, Apollo, ElevenLabs, etc.) through xpay, all calls bill to one account with one invoice.
• Instant activation — Start using Kokoro within seconds of signing up. No API key exchanges, no waiting for provider approval.