Boosting the Predictive Accurary of Singer Identification Using Discrete Wavelet Transform For Feature Extraction

TL;DR

This study enhances singer identification accuracy by introducing Discrete Wavelet Transform (DWT) for feature extraction, combined with SVM, outperforming traditional MFCC methods on a small dataset.

Contribution

The paper introduces the novel use of DWT for singer identification, demonstrating its effectiveness over MFCC in this context.

Findings

DWT with db4 achieved 83.96% accuracy.

DWT outperformed MFCC in feature extraction.

Support Vector Machine yielded the best results.

Abstract

Facing the diversity and growth of the musical field nowadays, the search for precise songs becomes more and more complex. The identity of the singer facilitates this search. In this project, we focus on the problem of identifying the singer by using different methods for feature extraction. Particularly, we introduce the Discrete Wavelet Transform (DWT) for this purpose. To the best of our knowledge, DWT has never been used this way before in the context of singer identification. This process consists of three crucial parts. First, the vocal signal is separated from the background music by using the Robust Principal Component Analysis (RPCA). Second, features from the obtained vocal signal are extracted. Here, the goal is to study the performance of the Discrete Wavelet Transform (DWT) in comparison to the Mel Frequency Cepstral Coefficient (MFCC) which is the most used technique in audio signals. Finally, we proceed with the identification of the singer where two methods have experimented: the Support Vector Machine (SVM), and the Gaussian Mixture Model (GMM). We conclude that, for a dataset of 4 singers and 200 songs, the best identification system consists of the DWT (db4) feature extraction introduced in this work combined with a linear support vector machine for identification resulting in a mean accuracy of 83.96%.