NUesc is IoT based environmental sound classification system for smart cities. The machine learning inference is deployed on Raspberry Pi to detect the events such as dog bark, car horn, gun shot and so on. The Raspberry Pi talks to the cloud server when the event is detected. The web application is used to manage the sounds for Raspberry Pi to listen to.
This is the final year project of Software Engineering from University of Newcastle, Australia. This project is currently in prototype version.
- Python 3
- Nodejs
- Raspberry Pi 3
- ReSpeaker 2-Mics Pi HAT
- UrbanSound8K Dataset
There are four components: model, nuesc-pi, server, and webapp. The detail documentation for each component can be found in each folder. The instructions for setup are as follows:
- First train the machine learning model. There are two models: SVM and KNN.
- Copy nuesc-pi into Rasbperry Pi.
- Copy trained model to nuesc-pi folder on Raspberry Pi. Make sure to name the model file as model.p
- Then, run the server from server folder.
- Finally, web application can be launched from webapp.
To setup Raspberry Pi and Respeaker 2-Mics Pi HAT, please follow http://wiki.seeedstudio.com/ReSpeaker_2_Mics_Pi_HAT/