MCP_server

# Speech Analysis MCP Server This project is a speech analysis server that uses the MCP (Model Context Protocol) architecture for automatic analysis of spoken English content. It supports features such as speech-to-text, grammar error analysis, vocabulary suggestions, logic suggestions, and original sentence comparison. ## Project Structure ``` . ├── models/ # Tool call encapsulation logic (e.g., tool_call.py) ├── myenv/ # Python virtual environment (not recommended for submission) ├── speech_audio/ # Reference or instructional audio files ├── tools/ # MCP tool collection │ ├── compareSentence.py # Compare original text with spoken differences │ ├── grammar.py # Grammar analysis tool │ ├── logic.py # Content logic suggestion tool │ ├── vocab.py # Vocabulary suggestion tool │ └── whisper.py # Speech-to-text processing ├── user_audio/ # User-uploaded audio files ├── main.py # MCP server main program ├── MCPtest.json # MCP test request template ├── requirements.txt # Python package installation list └── README.md # Project documentation ``` ## Feature Description - **Speech-to-Text**: Processes audio files using the OpenAI Whisper model - **Grammar Check**: Analyzes grammar errors - **Vocabulary Suggestions**: Provides better vocabulary choices - **Logic Suggestions**: Feedback on sentence structure and logic - **Sentence Comparison**: Highlights discrepancies with the original text ## Installation Steps 1. Clone the project: ```bash git clone https://github.com/YunYue1029/MCP_server cd MCP_server ``` 2. Create a virtual environment and install dependencies: ```bash python -m venv myenv source myenv/bin/activate # Windows: myenv\Scripts\activate pip install -r requirements.txt ``` 3. Start the MCP server: ```bash uvicorn main:app --reload ``` ## API & Testing Method You can use `main.py` with POST requests to trigger tool analysis; please refer to `MCPtest.json` for examples. ## Supported Audio Formats - `.mp3`, `.wav` - Test audio files can be placed in the `user_audio/` directory.