# ditto.audio

Gives your coding agent text-to-speech: it browses voices, synthesizes audio and hands back a download link.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 1
- Author: ditto-audio
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/ditto-audio/ditto-mcp
- Updated: 2026-10-07
- HTML: https://affpapa.org/en/nevibecode/mcp/dittoaudio

## What it does

The server lives at `https://mcp.ditto.audio` and speaks Streamable HTTP. On the first connection your client asks you to sign in to a ditto.audio account, so there is no key to paste. That connection shows up under Account -> API keys as a connected app, spends the same monthly allowance as your keys, and you can disconnect it from the same page.

Seven tools ship with it. `list_voices` walks the catalogue and your custom voices; `list_languages` lists language codes and which engines cover them. `text_to_speech` turns text into a download URL — MP3 unless you ask for FLAC, WAV or something else. Larger jobs go through `create_speech_batch`, which queues up to 1000 texts, with `get_speech_batch`, `get_speech_batch_results` and `cancel_speech_batch` for progress, links and stopping a run. Cancelling keeps whatever already finished.

Headless setups such as CI or a server skip the sign-in and pass a key from your API keys page as `Authorization: Bearer ditto_sk_live_...`. Audio arrives as a short-lived link rather than bytes in the conversation, so nothing heavy piles up in the chat history.

## Requirements
- Remote server: `https://mcp.ditto.audio`

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http tts https://mcp.ditto.audio
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add dittoaudio --url https://mcp.ditto.audio
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "dittoaudio": {
            "url": "https://mcp.ditto.audio"
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "dittoaudio": {
            "type": "http",
            "url": "https://mcp.ditto.audio"
        }
    }
}
```

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http dittoaudio https://mcp.ditto.audio
```

## When to use it
- synthesize a line of text and get an MP3 link
- check which voices the catalogue offers
- find out which engines speak a given language
- queue a large set of texts for synthesis
- stop a batch that is already running

## Example prompts
- List the ditto.audio voices and pick a female one for English
- Synthesize this text and give me the download link
- Queue every line in `script.txt` and send me the links when they are ready

## Similar
- [ttscn](https://affpapa.org/en/nevibecode/skills/ttscn.md) — skill · 1600 installs
- [tts](https://affpapa.org/en/nevibecode/skills/marswaveai-tts.md) — skill · 1400 installs
- [elevenlabs-tts](https://affpapa.org/en/nevibecode/skills/elevenlabs-tts.md) — skill · 1300 installs
- [speakturbo-tts](https://affpapa.org/en/nevibecode/skills/speakturbo-tts.md) — skill · 1000 installs
- [eachlabs-voice-audio](https://affpapa.org/en/nevibecode/skills/eachlabs-voice-audio.md) — skill · 308 installs
- [qianwen-audio-tts](https://affpapa.org/en/nevibecode/skills/qianwen-audio-tts.md) — skill · 5600 installs

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md