# TTS MCP Server

> TTS MCP Server is a freemium MCP server for design and media. Turn scripts into voiceover audio with matching SRT and VTT subtitles. It runs locally on your machine and is started by your AI client.

- **Kind:** MCP server
- **Pricing:** Freemium
- **Platforms:** CLI, API, macOS, Windows, Linux
- **Website:** https://miragea.ai
- **Repository:** https://github.com/rty2345rr/miragea-voice-mcp
- **MCP Registry name:** io.github.RTY2345RR/tts-mcp-server
- **Categories:** design and media, AI models and LLM tooling
- **Listing updated:** 2026-11-14

## About

One call can take a script file and write an audio narration plus .srt and .vtt captions side by side, ready for video editors such as CapCut, Premiere Pro or DaVinci Resolve. Caption timing comes from speech recognition on the generated audio, but the server puts back your exact script wording so brand names are not mis-spelled by the recogniser.

Beyond narration it can clone a voice from a reference audio URL, use voices saved in your Miragea account and transcribe local audio or video into captions. Output formats include MP3, WAV, FLAC, Opus, AAC and PCM, and caption length is adjustable for vertical short-form video.

The server runs locally over stdio (install with uvx, pip or Docker) and needs a MIRAGEA_API_KEY. It is MIT licensed, while speech generation runs on the Miragea service, whose plans meter audio credits and characters per request.

## Features

- **Script to voiceover:** One tool call writes narration audio and caption files together.
- **Script-faithful captions:** Keeps recognised timing but restores your exact wording.
- **Voice cloning:** Clones a voice from a reference audio URL; Miragea cites 5 to 10 seconds of sample on its site.
- **Subtitle formats:** Writes both SRT and WebVTT files.
- **Local transcription:** Transcribes local audio or video files into captions.
- **Several audio formats:** Exports MP3, WAV, FLAC, Opus, AAC or PCM.

## Good for

- Add a voiceover and captions to a YouTube Short
- Narrate a course or product explainer from a text script
- Dub a character with a cloned voice
- Generate captions for existing audio or video files

## Install

### Claude Code (Terminal)

```sh
claude mcp add tts-mcp-server --env MIRAGEA_API_KEY='<MIRAGEA_API_KEY>' -- uvx tts-mcp-server
```

### Claude Desktop (claude_desktop_config.json)

```json
{
  "mcpServers": {
    "tts-mcp-server": {
      "command": "uvx",
      "args": [
        "tts-mcp-server"
      ],
      "env": {
        "MIRAGEA_API_KEY": "<MIRAGEA_API_KEY>"
      }
    }
  }
}
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "tts-mcp-server": {
      "command": "uvx",
      "args": [
        "tts-mcp-server"
      ],
      "env": {
        "MIRAGEA_API_KEY": "<MIRAGEA_API_KEY>"
      }
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "tts-mcp-server": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "tts-mcp-server"
      ],
      "env": {
        "MIRAGEA_API_KEY": "<MIRAGEA_API_KEY>"
      }
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "tts-mcp-server": {
      "command": "uvx",
      "args": [
        "tts-mcp-server"
      ],
      "env": {
        "MIRAGEA_API_KEY": "<MIRAGEA_API_KEY>"
      }
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "tts-mcp-server": {
      "command": "uvx",
      "args": [
        "tts-mcp-server"
      ],
      "env": {
        "MIRAGEA_API_KEY": "<MIRAGEA_API_KEY>"
      }
    }
  }
}
```

## Tools

- `miragea_script_to_voiceover`: Generates narration from a script and writes audio plus SRT and VTT captions.

## FAQ

### What do I need to run it?

A MIRAGEA_API_KEY from the Miragea workspace; install with uvx tts-mcp-server, pip or Docker.

### Which clients does it support?

The README lists Claude Desktop, Claude Code, Cursor, OpenAI Codex, VS Code and Windsurf.

### Is the server open source?

Yes, it is MIT licensed, but generation uses the paid Miragea service.

### What was it called before?

miragea-voice-mcp; that command still works as an alias.

### Is TTS MCP Server free?

TTS MCP Server has a free plan, with paid plans for more usage or features.

### How do I add TTS MCP Server to Claude Code?

Run this command in your terminal: claude mcp add tts-mcp-server --env MIRAGEA_API_KEY='<MIRAGEA_API_KEY>' -- uvx tts-mcp-server

### What are the best alternatives to TTS MCP Server?

The closest alternatives to TTS MCP Server are HyperFrames by HeyGen, ffmpeg-skill, Claude Code Ultimate Guide, AvatarSay — Fictional Character & Talking Video MCP and ChartAnt.

## Alternatives

- [HyperFrames by HeyGen](https://ai-powered.com/mcp/hyperframes-by-heygen): HyperFrames by HeyGen is a freemium MCP server for design and media.
- [ffmpeg-skill](https://ai-powered.com/mcp/ffmpeg-skill): ffmpeg-skill is an open-source MCP server for design and media.
- [Claude Code Ultimate Guide](https://ai-powered.com/mcp/claude-code-ultimate-guide): Claude Code Ultimate Guide is an open-source MCP server for ai models and llm tooling.
- [AvatarSay — Fictional Character & Talking Video MCP](https://ai-powered.com/mcp/avatarsay-fictional-character-and-talking-video-mcp): AvatarSay — Fictional Character & Talking Video MCP is a freemium MCP server for design and media.
- [ChartAnt](https://ai-powered.com/mcp/chartant): ChartAnt is a freemium MCP server for design and media.
- [VfxSmith](https://ai-powered.com/mcp/vfxsmith): VfxSmith is a freemium MCP server for design and media.
- [CalderFlow](https://ai-powered.com/mcp/calderflow): CalderFlow is a free MCP server for design and media.
- [Hourlore](https://ai-powered.com/mcp/hourlore): Hourlore is a freemium MCP server for design and media.

---
Source: https://ai-powered.com/mcp/tts-mcp-server
