Alog - AI-Powered Wearable Device

Alog is an AI-powered wearable device application that processes audio input using machine learning models to detect and classify sounds and speech in real-time.

Quick Start

To get started with Alog, follow these simple steps:

1. Clone the repository:

   git clone https://github.com/yourusername/alog.git
   cd alog

2. Run the installation script:

   sudo ./install.sh

The installation script will:

• Install all necessary system dependencies
• Set up Python virtual environment
• Install Python packages
• Build the frontend
• Configure automatic login and log viewing
• Set up and start the system service

Accessing Logs

There are three ways to view the application logs:

1. Direct Display (HDMI):

   • Connect a display via HDMI
   • The logs will automatically appear on TTY1 after boot
   • No login required - the system automatically logs in and shows logs

2. SSH Access:

   ssh khadas@your-device-ip
   journalctl -f -u alog.service --no-hostname --output=cat

3. Log File:

   tail -f /home/khadas/alog/logs/app.log

Web Interface

Access the web interface at http://your-device-ip:5000

Configuration

Log Levels

By default, the application shows EVENT, SPEECH, and ERROR level logs. To modify this:

1. Edit the service configuration:

   sudo nano /etc/systemd/system/alog.service

2. Modify the ENABLED_LOG_LEVELS environment variable:

   Environment="ENABLED_LOG_LEVELS=EVENT,SPEECH,ERROR,DEBUG"

3. Restart the service:

   sudo systemctl restart alog.service

Available log levels: DEBUG, INFO, EVENT, SPEECH, WARNING, ERROR, CRITICAL

Development

Project Structure

alog/
├── audio_chunks/       # Stored audio segments
├── backend/           # Python backend
│   ├── api/          # REST API endpoints
│   ├── models/       # Data models
│   └── services/     # Core services
├── frontend/         # React frontend
│   └── src/         # Frontend source code
└── models/          # ML models
    ├── vosk/        # Speech recognition
    └── yamnet/      # Sound classification

Running in Development Mode

For development, you can run the application manually:

1. Start the backend:

   source venv/bin/activate
   python backend/app.py

2. Start the frontend (in a separate terminal):

   cd frontend
   npm start

Troubleshooting

1. No logs appearing on HDMI display:

   • Check if the service is running: systemctl status alog.service
   • Verify TTY1 configuration: cat /etc/systemd/system/[email protected]/override.conf

2. Service fails to start:

   • Check logs: journalctl -u alog.service -n 50
   • Verify permissions: ls -la ~/alog
   • Check Python environment: ~/alog/venv/bin/python -V

3. No audio detection:

   • Verify audio device: arecord -l
   • Check permissions: groups khadas