Speech MCP

🚀 Speech MCP Documentation

This documentation provides a comprehensive guide to the Speech MCP, including installation, configuration, features, usage examples, troubleshooting, and technical details.

🚀 Quick Start

To quickly get started with Speech MCP, follow the installation steps below and start using it right away.

✨ Features

• Text-to-Speech (TTS): Supports multiple TTS engines, including the default pyttsx3 and the optional Kokoro TTS.
• Speech-to-Text (STT): Uses the faster-whisper engine for local speech recognition, supporting various audio formats.
• Multilingual Support: Allows for mixed input and recognition of Chinese and English.
• Customizable Voice Models: Users can download and add additional voice models.

📦 Installation

Compatibility

• Operating Systems: Supports Windows, macOS, and Linux (Ubuntu 20.04+).
• Python Version: Python 3.6+.

Installation Steps

1. Install dependencies using pip:

   pip install speech-mcp
   

2. Download and install the PortAudio library:
   • macOS:

     brew install portaudio
     export LDFLAGS="-L/usr/local/lib"
     export CPPFLAGS="-I/usr/local/include"
     pip install pyaudio
     

   • Linux (e.g., Ubuntu):

     sudo apt-get update && sudo apt-get install portaudio19-dev python3-dev
     pip install pyaudio
     

📚 Documentation

Configuration Guide

Default Configuration File Path

• User Configuration File: ~/.config/speech-mcp/config.json

Environment Variable Settings

• SPEECH_MCP_TTS_VOICE: Sets the default Text-to-Speech (TTS) engine.
• SPEECH_MCP_STT_MODEL: Sets the default Speech-to-Text (STT) model.

Function Overview

Text-to-Speech (TTS)

Supports multiple TTS engines:

1. Default Engine: pyttsx3
   • Uses the system's built-in voices.
   • Supports multiple languages and voice styles.
2. Optional Engine: Kokoro TTS (requires additional installation)
   • High-quality neural network-driven speech synthesis.
   • Supports multiple languages and voice styles.

Speech-to-Text (STT)

• Uses the faster-whisper engine for local speech recognition.
• Supports multiple audio formats, including WAV, MP3, and FLAC.

Troubleshooting

Common Issues

1. Audio device not detected
   • Ensure that the microphone is properly connected and recognized by the system.
   • Specify the specific device index in the code (if there are multiple devices).
2. PortAudio installation failed
   • Check the system logs or reinstall the relevant dependencies.

Update Log

For detailed update records, please refer to the Changelog in the project documentation.

💻 Usage Examples

Basic Usage

1. Start the Speech MCP:

   speech-mcp start
   

2. Control the device via voice:
   • Say "Let's start using voice" or a similar command to start the voice mode.
   • The system will automatically switch to the voice input interface.

Advanced Usage

• Multilingual Support: Supports mixed input and recognition of Chinese and English.
• Custom Voice Models: Users can download and add additional voice models.

🔧 Technical Details

Internal Working Principle

• Uses the faster-whisper engine for speech recognition.
• Implements the Text-to-Speech function through pyttsx3 or Kokoro TTS.

Performance Optimization Suggestions

1. Hardware Acceleration
   • Use GPU acceleration (e.g., devices supporting CUDA) to improve inference speed.
2. Model Selection
   • Select lightweight or high-performance voice models according to your needs.