"VoxLabs: Real-Time Vocal Trainer" by Madhav Danturthi and Aidan Galia

Date of Award

2024

Document Type

Thesis

Publisher

Santa Clara : Santa Clara University, 2024

Department

Electrical and Computer Engineering

First Advisor

Sally Wood

Second Advisor

Andy Wolfe

Abstract

With the rapid evolution of society, the barriers to entry into the world of music are more constricting than ever. The impacts of race, age, gender, and language on the quality of musical education and the accessibility to education in general is growing more and more. That is why it is more important than ever to develop a new accessible method of vocal training focused around improving craft through gamification and rewards, all while inspiring a next generation of ready to learn musicians.

VoxLabs is a project created to help music lovers learn how to sing through an entertaining yet challenging environment designed to enhance users’ singing abilities through real-time corrective feedback, a plethora of musical lessons, and many fun games to play along the way. Younger audiences such as elementary school students, as well as people with a love for music, need our vocal trainer because it is a more cost efficient and enjoyable way to learn how to sing both professionally and casually.

Realizing that singing in pitch is one of the most challenging yet fundamental tasks of singing, we started our development based around matching and reproducing a single tone, progressing to reproducing a sequence of tones, to finally reproducing an entire song. However, we decided to reduce the barrier of singing in pitch, creating a project that allows users to gain practice and experience at their own pace in reproducing tones with ease, through interactive visual lessons and feedback.

By making use of Signal Processing and STMicroelectronics’ STM3240-G EVAL board, we created an audio processing algorithm that takes a user inputted audio which is then analyzed to obtain the fundamental frequencies sung by the user. We then compare these frequencies to a given control audio to give real-time feedback to the user, helping the user improve their singing abilities. With the time given, we focused on creating an effective algorithm and creating a proof of concept for the hardware aspect of our senior design project, creating a hardware system with the capability to test the effectiveness of our created algorithm.

Download

Share

COinS