mcp-tts-gemini mcp

v1.0.0 · MCP Tool · audio · registry.pascalai.org

Text-to-speech via Google Gemini TTS API (gemini-2.5-flash-preview-tts). High-quality natural voices. Multi-speaker support. Returns WAV audio (base64 or file). Auth via apiKey or GEMINI_API_KEY env var.

ttsgeminigooglespeechaiaudiomulti-speaker

Input Parameters

ParameterTypeDescription
textrequired string Text to convert to speech.
voiceoptional string Voice name: Puck (default), Aoede, Charon, Fenrir, Kore, Zephyr, Achernar, Leda, Orus, etc. Multi-speaker: "Anya=Kore, Liam=Puck". Default: Puck.
modeloptional string Gemini TTS model (default: gemini-2.5-flash-preview-tts). Default: gemini-2.5-flash-preview-tts.
outputPathoptional string Save audio to this file path (e.g. C:\speech.wav). Returns base64 WAV if empty.
formatoptional string Output format: wav (only supported format). Default: wav.
apiKeyoptional string Gemini API key (or GEMINI_API_KEY env var).
baseUrloptional string Base URL override (default: https://generativelanguage.googleapis.com/v1beta). Default: https://generativelanguage.googleapis.com/v1beta.
audioProfileoptional string Audio style instructions (e.g. "Speak in a calm, friendly tone with slight enthusiasm").
sceneoptional string Scene context to guide the voice performance.
directorsNotesoptional string Director's notes to shape the delivery style.

Output Fields

FieldTypeDescription
ok boolean
audio_b64 string Base64-encoded WAV audio (when outputPath not set).
output_path string File path where audio was saved.
format string
size_bytes integer
error string

Examples

Basic speech synthesis with Puck voice

// Input
{
  "text": "Hello! I'm your AI assistant. How can I help you today?",
  "voice": "Puck",
  "apiKey": "AIza..."
}

// Output
{
  "ok": true,
  "audio_b64": "UklGRg...",
  "format": "wav",
  "size_bytes": 115200
}

Multi-speaker dialogue

// Input
{
  "text": "Anya: Good morning! Liam: Good morning to you too!",
  "voice": "Anya=Kore, Liam=Puck",
  "apiKey": "AIza...",
  "outputPath": "C:\\audio\\dialogue.wav"
}

// Output
{
  "ok": true,
  "output_path": "C:\\audio\\dialogue.wav",
  "format": "wav",
  "size_bytes": 230400
}

Expressive narration with director mode

// Input
{
  "text": "In a world where silence speaks louder than words...",
  "voice": "Aoede",
  "audioProfile": "Deep, cinematic narrator voice with gravitas",
  "directorsNotes": "Slow and deliberate pace, dramatic pauses",
  "apiKey": "AIza..."
}

// Output
{
  "ok": true,
  "audio_b64": "UklGRg...",
  "format": "wav",
  "size_bytes": 192000
}

Install & Discovery

Install

ppm install mcp-tts-gemini

Get JSON Schema

GET /v1/packages/mcp-tts-gemini/1.0.0/schema

Discover by keyword

GET /v1/mcp/discover?q=tts
Discovery hint: Install with ppm install mcp-tts-gemini or invoke remotely via POST /v1/invoke/mcp-tts-gemini on the MCP Service.