
Analyzing Tennis Matches Based on Audio
EECS 351 Fall 2020 Final Project
Erik Sangeorzan, Jason Hu, Max Baer, Pavel Mazza, Po-Hsun Wu
Page Contents
Quick Overview
The goal of our project is to create software that takes in an audio clip of a tennis match and analyzes the audio in various ways relevant to the game in order to draw conclusions pertaining to the status of the match.
Given how the crowd is mostly silent while the ball is in play, we will be able to identify the sounds that pertain to how the game is scored, such as a player hitting the ball across the court or the ball hitting the net, thereby ending the point. Our main goals include detecting the beginning and end of a point, estimating the winner of the point, and categorizing the various sounds heard throughout the match, both during and between points.
​
We seek to achieve our goals primarily though audio processing and filtering in MATLAB, especially using low/high pass filters and the FFT algorithm. Additional ideas such as event categorization, though challenging for the scope of this project, involve designing filters that are not time-invariant.
​
Check out the Page Contents section above to get a summary of what each page on our website covers.