A comparative study of pitch extraction algorithms on a large variety of singing sounds

Onur Babacan (1) , Thomas Drugman (1) , Nicolas d'Alessandro (1) , Nathalie Henrich Bernardoni (2) , T. Dutoit (1)

1 Circuit Theory and Signal Processing Laboratory
2 GIPSA-SLD - GIPSA - Systèmes Linguistiques et Dialectologie

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 pitchextraction algorithms, depending on singer category and laryngeal mechanism. Finally, we analyze their robustness to reverberation.

Mots clés: singing analysis/synthesis; pitch extraction

Mots clés: analyse de la voix chantée; extraction de la hauteur; synthèse

Domaines: Traitement du signal et de l'image [eess.SP]; Traitement du signal et de l'image [eess.SP]; Acoustique [physics.class-ph]; Acoustique [physics.class-ph]; Sciences de l'information et de la communication; Linguistique

Onur Babacan, Thomas Drugman, Nicolas d'Alessandro, Nathalie Henrich Bernardoni, T. Dutoit. A comparative study of pitch extraction algorithms on a large variety of singing sounds. ICASSP 2013 - 38th IEEE International Conference on Acoustics, Speech and Signal Processing, May 2013, Vancouver, Canada. pp.1-5. ⟨hal-00923967⟩