Does the Ear of the Eigenface Hear the Eigensound?
[Case Study] Application of PCA in Music Classification.
In this project, we explore the classification of audio files, specifically music pieces ranging from the classical solo piano (Chopin, Mozart, and Liszt), pop music (Adele and Billboard top 30 pop songs) to jazz. Considering the immense feature-space of our data (220,500 features per 5-second sample), exploitation of eigensounds has formed the principal components of our data preprocessing. Specifically, considering low-throughput networks that are still standard in many places worldwide and the amount of data available and needed for training a useful model, we have focused on minimizing the amount of sound data (quality and decomposition) needed for fast and accurate classification. You can access the full paper here (https://sadatian.io/projects/Eigensounds.pdf).
Click here to view the embedded file.