Skip to content

An integrated AI-driven system that combines computer vision, voice processing, and sensor data to create an intelligent and responsive environment.

License

Notifications You must be signed in to change notification settings

ducroq/sensAIhub-main

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

sensAIhub-main

An integrated AI-driven system that combines computer vision, voice processing, and sensor data processing to create an intelligent and responsive environment.

Overview

SensAIHub leverages various sensors and AI technologies to enable data-driven user interactions. By processing multiple streams of sensory data in real-time, the system can respond intelligently to user presence, actions, and inputs, creating a highly responsive and adaptive environment. Designed to run primarily on a Raspberry Pi, with additional processing capabilities provided by a backend GPU server, SensAIHub aims to push the boundaries of human-computer interaction.

Key Concepts

  • Data-Driven Interaction: Utilizes real-time sensor data to inform and enhance user interactions
  • Multimodal Sensing: Combines visual, auditory, and environmental data for comprehensive user understanding
  • Adaptive Response: Tailors system behavior based on analyzed user data and context
  • Edge Computing: Balances local processing with cloud capabilities for efficient operation

Features

  • Computer Vision: Face detection, landmark tracking, head orientation, and gaze direction estimation using Google MediaPipe
  • Voice Processing: Wake-word detection, speaker diarization, voice pitch estimation, and speech-to-text conversion
  • Conversational AI: Integration with a small LLM (e.g., Llama or Geitje) for natural language interaction
  • Presence Detection: Utilizes radar sensors for user presence, breathing rate, and potentially heart rate monitoring
  • Distributed Processing: Local processing on Raspberry Pi with heavy computational tasks offloaded to a GPU server
  • Remote Development: Supports remote development and debugging via Tailscale network

System Architecture

  • Edge Device: Raspberry Pi 4
  • Auxiliary MCU: Raspberry Pi Zero (for low-level sensor integration)
  • Backend: GPU server for intensive processing tasks
  • Network: Tailscale VPN for secure remote access

Development Setup

  1. Set up Raspberry Pi with Raspberry Pi OS
  2. Install Tailscale on all devices (Pi, development machine, server)
  3. Configure VSCode with Remote-SSH extension for development
  4. Clone this repository and its submodules

Detailed setup instructions can be found in the docs/setup.md file.

Contributing

We welcome contributions! Please see our Contributing Guidelines for more details.

License

This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

Acknowledgements

This project makes use of several open-source libraries and tools. We thank all the contributors to these projects.

Contact

For any queries regarding this project, please open an issue in this repository.

About

An integrated AI-driven system that combines computer vision, voice processing, and sensor data to create an intelligent and responsive environment.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages