Inspiration: We decided on working with sound since our group had a lot of interest in music and experience with sound frequencies
What it does: It is a music creation studio that converts only vocal inputs into any number of other instrument outputs (i.e. Sing into mic and choose trumpet to get a trumpet solo)
How we built it: The program was built on Python. We take the vocal input and parse it into frequencies and time when the tune is played. Then use frequency and time input in librosa library to change the frequency and length of the stock sound files from different instruments. GUI is a program that takes user input and converts it into selected instrument sound with the option of mixing multiple sound files and plays it.
Challenges We ran into : Recording user audio using multi threading was complicated because we would want to return the sound file as we kill the thread. Calculating the white noise between vocal inputs and adding that to instrument audio. Also dealing with audio that involves slurring is challenging as there is no natural breakpoints between notes.
Accomplishments that we're proud of : Manipulating different frequencies to get specific notes on a scale and adding tabs for guitar sound. Creating the entire library for every instrument using only one audio file per instrument to keep it lightweight
What we learned: Using tkinter library for GUI creation, combining audiofiles using pyaudio, manipulating audio in Fourier Graph software, mathematics equations to find frequency differences in half steps, a lot of python