https://github.com/ashish-soni08/ai-speech-coach-omi-mcp
An AI-powered tool providing personalized feedback and guidance to improve articulation and speaking skills. Built for the MCP Hackathon at AGI House.
https://github.com/ashish-soni08/ai-speech-coach-omi-mcp
Last synced: 2 months ago
JSON representation
An AI-powered tool providing personalized feedback and guidance to improve articulation and speaking skills. Built for the MCP Hackathon at AGI House.
- Host: GitHub
- URL: https://github.com/ashish-soni08/ai-speech-coach-omi-mcp
- Owner: Ashish-Soni08
- License: apache-2.0
- Created: 2025-03-15T19:46:25.000Z (3 months ago)
- Default Branch: main
- Last Pushed: 2025-03-15T21:26:46.000Z (3 months ago)
- Last Synced: 2025-03-15T22:25:27.031Z (3 months ago)
- Size: 5.86 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# AI-Speech-Coach
An AI-powered tool providing personalized feedback and guidance to improve articulation and speaking skills. It analyzes speech patterns from OMI device conversations, providing end-of-day reports with specific improvement suggestions.
## Sparking the Idea: Video Inspiration
[](https://www.youtube.com/watch?v=2FTx7DV7sv8)
## Features
- **Filler Word Detection**: Identifies and counts filler words like "um," "uh," "like," etc.
- **Speech Pace Analysis**: Measures words per minute and speaking rhythm
- **Vocabulary Assessment**: Evaluates word variety and suggests improvements
- **Confidence Scoring**: Analyzes speaking confidence based on multiple metrics
- **Personalized Suggestions**: Provides actionable recommendations for improvement
- **Progress Tracking**: Monitors improvement over time with historical data
- **End-of-Day Analysis**: Scheduled job at 7 PM to process the day's conversations
- **MCP Integration**: Speech analysis tools accessible through Model Context Protocol
## Architecture
- **Backend**: FastAPI with async PostgreSQL database
- **Frontend**: Next.js with shadcn/ui components and responsive design
- **Analysis Engine**: Speech analysis components for filler words, pace, and vocabulary
- **MCP Server**: Provides speech analysis tools via Model Context Protocol
- **OMI Integration**: Real-time transcript and audio processing
## Setup & Installation
### Quick Start with Docker
1. Clone the repository:
```bash
git clone https://github.com/yourusername/AI-Speech-Coach-OMI-MCP.git
cd AI-Speech-Coach-OMI-MCP
```
2. Use the start script to launch the application:
```bash
chmod +x start.sh
./start.sh prod
```
3. Access the application:
- Backend API: http://localhost:8000
- Frontend Dashboard: http://localhost:3000
4. To stop the application:
```bash
./start.sh stop
```
### Manual Setup
#### Backend
1. Create a virtual environment and install dependencies:
```bash
cd ai-speech-coach
pip install -r requirements.txt
```
2. Set up environment variables in `.env` file:
```
DATABASE_URL=postgresql+asyncpg://postgres:postgres@localhost/speech_coach
PORT=8000
HOST=0.0.0.0
DEBUG=True
MCP_TRANSPORT=stdio
```
3. Initialize the database and start the server:
```bash
python setup_database.py
python main.py
```
#### Frontend
1. Install dependencies:
```bash
cd frontend
npm install
```
2. Create `.env.local` file:
```
NEXT_PUBLIC_API_URL=http://localhost:8000/api
```
3. Start the development server:
```bash
npm run dev
```
## Key Components
### Backend API Endpoints
- `POST /api/transcript/analyze`: Analyze transcript segments for speech patterns
- `GET /api/transcript/history/{user_id}`: Get historical analysis for a user
- `GET /api/transcript/statistics/{user_id}`: Get aggregated statistics for a user
- `POST /api/audio/upload`: Upload audio for analysis
- `POST /trigger-analysis`: Manually trigger end-of-day analysis
### MCP Tools
- `analyze_text`: Comprehensive speech analysis
- `detect_filler_words`: Identify and count filler words
- `generate_improvement_suggestions`: Get personalized improvement suggestions
### Dashboard Features
- Speech metrics visualization with charts and cards
- Historical progress tracking
- Personalized improvement suggestions panel
- Audio upload for analysis
## Project Structure
- `/ai-speech-coach`: Backend FastAPI application
- `/analyzer`: Speech analysis algorithms
- `/api`: API routes and services
- `/mcp`: MCP server implementation
- `/models`: Database models and schemas
- `/frontend`: Next.js dashboard application
- `/src/components`: React components
- `/src/lib`: Utilities and API service
- `/src/app`: Page components and routing
## Contributing
1. Fork the repository
2. Create a feature branch (`git checkout -b feature/your-feature`)
3. Commit your changes (`git commit -m 'Add feature'`)
4. Push to the branch (`git push origin feature/your-feature`)
5. Open a Pull Request
**Built for the MCP Hackathon at AGI House.**