An MCP server that provides seamless integration between Fish Audio's Text-to-Speech API and LLMs like Claude, delivering natural language speech synthesis capabilities with voice cloning, multilingual support, and streaming features.

Installation

NPX (Direct Run)

npx @alanse/fish-audio-mcp-server

Global Installation via NPM

npm install -g @alanse/fish-audio-mcp-server

From Source

git clone https://github.com/da-okazaki/mcp-fish-audio-server.git
cd mcp-fish-audio-server
npm install
npm run build
npm run dev

Configuration

Single Voice Mode

{
  "mcpServers": {
    "fish-audio": {
      "command": "npx",
      "args": ["-y", "@alanse/fish-audio-mcp-server"],
      "env": {
        "FISH_API_KEY": "your_fish_audio_api_key_here",
        "FISH_MODEL_ID": "speech-1.6",
        "FISH_REFERENCE_ID": "your_voice_reference_id_here",
        "FISH_OUTPUT_FORMAT": "mp3",
        "FISH_STREAMING": "false",
        "FISH_LATENCY": "balanced",
        "FISH_MP3_BITRATE": "128",
        "FISH_AUTO_PLAY": "false",
        "AUDIO_OUTPUT_DIR": "~/.fish-audio-mcp/audio_output"
      }
    }
  }
}

Multiple Voices Mode

{
  "mcpServers": {
    "fish-audio": {
      "command": "npx",
      "args": ["-y", "@alanse/fish-audio-mcp-server"],
      "env": {
        "FISH_API_KEY": "your_fish_audio_api_key_here",
        "FISH_MODEL_ID": "speech-1.6",
        "FISH_REFERENCES": "[{'reference_id':'id1','name':'Alice','tags':['female','english']},{'reference_id':'id2','name':'Bob','tags':['male','japanese']},{'reference_id':'id3','name':'Carol','tags':['female','japanese','anime']}]",
        "FISH_DEFAULT_REFERENCE": "id1",
        "FISH_OUTPUT_FORMAT": "mp3",
        "FISH_STREAMING": "false",
        "FISH_LATENCY": "balanced",
        "FISH_MP3_BITRATE": "128",
        "FISH_AUTO_PLAY": "false",
        "AUDIO_OUTPUT_DIR": "~/.fish-audio-mcp/audio_output"
      }
    }
  }
}

Available Tools

Tool	Description
fish_audio_tts	Generates speech from text using Fish Audio's TTS API with support for custom voices, streaming...
fish_audio_list_references	Displays all configured voice references with their IDs, names, and tags

Features

• High-quality TTS with modern speech synthesis
• Streaming support for real-time audio streaming and low-latency applications
• Multiple voice support with custom voice models via reference ID
• Intelligent voice selection by ID, name, or tags
• Voice library management for configuring and managing multiple voice references
• Support for multiple audio formats (MP3, WAV, PCM, Opus)
• Voice cloning capabilities for creating custom voice models
• Multilingual support including English, Japanese, Chinese, and more
• Real-time WebSocket streaming with live playback
• Flexible configuration via environment variables

Environment Variables

Required

• FISH_API_KEY - Your Fish Audio API key

Optional

• FISH_MODEL_ID - TTS model to use (s1, speech-1.5, speech-1.6)
• FISH_REFERENCE_ID - Default voice reference ID (single reference mode)
• FISH_REFERENCES - Multiple voice references in JSON format
• FISH_DEFAULT_REFERENCE - Default reference ID when using multiple references
• FISH_OUTPUT_FORMAT - Default audio format (mp3, wav, pcm, opus)
• FISH_STREAMING - Enable streaming mode (HTTP/WebSocket)
• FISH_LATENCY - Latency mode (normal, balanced)
• FISH_MP3_BITRATE - MP3 bitrate (64, 128, 192)

Usage Examples

Generate speech with the text 'Hello, world! Welcome to Fish Audio TTS.'

Generate speech with voice model xyz123 saying 'This is a custom voice test'

Use Alice's voice to say 'Hello from Alice'

Generate Japanese speech with the text 'こんにちは' in an anime voice

What voices are available?

Notes

Requires a Fish Audio API key. Supports up to 10,000 characters per TTS request. Voice selection priority: reference_id > reference_name > reference_tag > default. Includes error handling for API limits, network issues, and invalid parameters.