A Comparative Study of Pitch Extraction Algorithms on a Large Variety of Singing Sounds

TL;DR

This study compares various state-of-the-art pitch extraction algorithms on a large, diverse singing voice database to evaluate their accuracy, robustness, and adaptability for singing voice analysis.

Contribution

It provides a comprehensive evaluation of existing pitch-tracking methods specifically adapted for singing voices, highlighting their strengths and limitations.

Findings

Algorithms vary in accuracy depending on singer category.

Reverberation affects the robustness of pitch extraction methods.

Adaptation of speech-based methods improves singing voice analysis.

Abstract

The problem of pitch tracking has been extensively studied in the speech research community. The goal of this paper is to investigate how these techniques should be adapted to singing voice analysis, and to provide a comparative evaluation of the most representative state-of-the-art approaches. This study is carried out on a large database of annotated singing sounds with aligned EGG recordings, comprising a variety of singer categories and singing exercises. The algorithmic performance is assessed according to the ability to detect voicing boundaries and to accurately estimate pitch contour. First, we evaluate the usefulness of adapting existing methods to singing voice analysis. Then we compare the accuracy of several pitch-extraction algorithms, depending on singer category and laryngeal mechanism. Finally, we analyze their robustness to reverberation.