Scientific Python Data Acoustic Modeling

This is a project for COP 2080, CS Problem Solving and Solution. We used the skills from this class to model the acoustics of a clap in an enclosed space.

About the Project

It's evident that some indoor spaces have better acoustics than others, this project deals with voice and audio intelligibility within these spaces. With the use of Python, we have created a tool to help improve audio intelligibility. For this project, we recorded sample audio from the Aula Magna, an indoor auditorium that has at least 1 second of reverb time.

Our first step was to create a clean and simple GUI that was user-friendly and easily understood. We used Tkinter, a standard GUI toolkit for Python to create the GUI that our user will fetch their desired audio sample from. We added a button that allows the user to access their files and select their audio file. The accepted file types are .mp3 and .wav, however, to manipulate and plot this data (this will be discussed later) we will need to convert the audio file to the .wav file format.

After selecting the audio sample and converting it, we will manipulate and plot the audio in several different ways. We will remove the metadata, isolate the audio, and plot it using SciPy. We will graph several different aspects of the audio file.

Installation

In order to use this project properly, you must pip-install these libraries:

• tkinter (to create the GUI)-------------------------------------------------------> In the terminal pip install tkinter
• pydub (used for converting and manipulating the audio)---------------------> In the terminal pip install pydub
• ffmpeg (pydub works because of this framework)-----------------------------> In the terminal pip install ffmpeg
• scipy (manipulation and plotting audio)----------------------------------------> In the terminal pip install scipy
• numpy (Aray Computing) ----------------------------------------> In the terminal pip install numpy

You must also add FFmpeg to your system's path if you don't already

1. Go to https://ffmpeg.org/download.html
2. Under "Get packages and executable files" select your Operating System (Linux, Windows, or MacOS)
3. There will be a dropdown menu for each OS, select a package suitable for your system and download it
4. Extract the folder and navigate to the executable (.exe file)
5. Copy the directory path
6. Open "System Properties"
7. Click on "Advanced Settings"
8. Click on "Environment Variables"
9. Find the "Path" variable and click "Edit"
10. Now click "New" and paste the directory path you copied in Step 5 and click "Ok"
11. Open a command prompt by typing "cmd" in the search box on your task bar
12. Once opened type "ffmpeg" to ensure it has been properly added to the path

Usage

Follow these steps to use our project:

1. Ensure that you have all the necessary libraries installed
2. Run the program (you should now see the GUI)
3. Click on the "Select Audio File" button and select your audio sample
4. You may then choose any of the buttons to view the plots of your audio sample