On fusion of timbre-motivated features for singing voice detection and singer identification

Timbre is the quality of sound which allows the ear to distinguish between musical sounds. In this paper, we study timbre effects in identification of singing voice segments in popular songs. Firstly, we identify between singing voice and instrumental segments in a song. Then, singing voice segments are further categorized according to their singer identity. Timbre-motivated effects are formulated by fusion of systems that use the features from vibrato, harmonic information and other features extracted using Mel and Log frequency scale filter banks. Statistical methods to select singing voice segments with high confidence measure are proposed for better performance in singer identification process. The experiments conducted on a database of 214 popular songs show that the proposed approach is effective.

1. INTRODUCTION

Singing voice detection (SingVD) and Singer identification (SingerID) are the two important problems in the area of music information retrieval (MIR). The problem of SingVD is formulated as follows: given a song, classify each segment of the song as to whether it is purely instrumental (nonvocal) segment, or a mixture of vocals with/without background instrumental (vocal) segment. For the problem of SingerID, it is formulated as identifying singer of vocal segments using a classifier such as Hidden Markov Models (HMM). Although considerable progress has been made, it remains a challenge to automate the SingVD and SingerID systems with high precision.