Safe and Sound

SafeAndSound
SafeAndSound Demo

Inspiration

While we were standing in the lobby for PolyHacks, one of us made a random loud noise and somehow, that lead to the inspiration of trying to reduce loud, unnecessary obnoxious noise. Living in the city, the omnipresent noise pollution is a problem that has become a norm in our urban society. In fact, recent studies effectively show that noise pollution can affect an individual’s mental and/or physical health.

What it does

SafeAndSound is a program that allows cities and communities to detect different noise levels in a given time frame (i.e. 24 hours) in order to be able to identify these peak noise pollution hours and tackle them.

How we built it

The program was built entirely using Python (PyAudio, Tkinter, Matplotlib) and our good friend ChatGPT. -PyAudio: Detect sound level (dB) -Matplotlib: Graph and visualize the varying sound levels over time -Tkinter: Display the information in a window for the user to see Our code displays a dynamic sound bar that changes depending on the audio level, a live measure of sound (dB), a first graph displaying the sound level over a time axis, and a second graph displaying the average sound level in a specified interval (e.g. 10 seconds) over a time axis. This second graph can be used by cities/communities who are running the program over e.g. 24 hours, so Safe and Sound would graph the average sound level recorded in every 20 minute interval (instead of plotting a 24 hours graph with every single data collected).

Challenges we ran into

Originally, the idea was to be able to build an AI that would be capable of distinguishing different types of sound (e.g. cars honking, humans talking, airplane flying, etc.), but with our skill level and the time constraint, we were not able to figure out how to do this. With this, we would have been able to provide users with more detailed and specific feedback (i.e. tell users exactly what needs to reduced).

Accomplishments that we're proud of

Despite the challenge, we took a risk with this project by stepping out of our comfort zone and working with sound, something that none of us were familiar with. We were also proud of having a half-decent UI considering that none of us have every working with Tkinter as well.

What we learned

PyAudio, Matplotlib, and Tkinter

What's next for Safe and Sound

As mentioned, it would be wonderful if we are able to upgrade Safe and Sound and be able to have it identify different sound in an environment. Our vision with this project is to have loud places, such as Montreal, be able to use this technology in different areas of the city in order to not only pin point when peak noise pollution happens but also what is causing it. This way, it would be much easier for cities to know what they are dealing with exactly and reduce noise pollution for the better of society.