Singer identification using perceptual features and cepstral coefficients of an audio signal from Indian video songs

Abstract

Singer identification is a difficult topic in music information retrieval because background instrumental music is included with singing voice which reduces performance of a system. One of the main disadvantages of the existing system is vocals and instrumental are separated manually and only vocals are used to build training model. The research presented in this paper automatically recognize a singer without separating instrumental and singing sounds using audio features like timbre coefficients, pitch class, mel frequency cepstral coefficients (MFCC), linear predictive coding (LPC) coefficients, and loudness of an audio signal from Indian video songs (IVS). Initially, various IVS of distinct playback singers (PS) are collected. After that, 53 audio features (12 dimensional timbre audio feature vectors, 12 pitch classes, 13 MFCC coefficients, 13 LPC coefficients, and 3 loudness feature vector of an audio signal) are extracted from each segment. Dimension of extracted audio features is reduced using principal component analysis (PCA) method. Playback singer model (PSM) is trained using multiclass classification algorithms like back propagation, AdaBoost.M2, k-nearest neighbor (KNN) algorithm, naïve Bayes classifier (NBC), and Gaussian mixture model (GMM). The proposed approach is tested on various combinations of dataset and different combinations of audio feature vectors with various Indian male and female PS's songs.

New Algorithm for the Development of a Musical Words Descriptor for the Artificial Composition of Oriental Music

“…In [32], the studies highlighted functionalities of SVM on a common songs database of 409 sounds of 16 groups, a comparison of SVM-based distinction Guo was established with other traditional methods, while proposing a new audio recovery criterion, called Boundary Deviation (DFB). [33] work on automated singer identification by distinguishing instrumental and singing sounds using audio information such as timbre parameters, pitch level, mel frequency cepstral coefficients (MFCC), linear predictive coefficients (LPC) and Indian video songs (IVS) audio signal loudness. In [34], an examination of the influence of texture choice on the identification of automated music genres and a novel K-Means-based texture filter aimed at distinguishing different sound textures in each album.…”

Robust singer identification of Indian playback singers

“…Saurabh et al [30] identified ten singers from North Indian classical music using timbre descriptors for noise-free studio recordings accompanied with continuous background music of tanpura and violin/harmonium/flute, etc., and achieved an accuracy of 58.33%. A technique of identifying singers using video songs from the Internet and CD for Indian playback singers using cepstral coefficients is presented in [31,32], however tested for a lower corpus. It is quite obvious, as the number of music items (songs and singers) increase, the performance of the system degrades due to noise and presence of more similar songs in the database [28,33].…”

A Survey on Content Based Video Retrieval Using Mpeg-7 Visual Descriptors

“…The HSV performed better than the HSL and YCbCr. Ka-Man Wong, Lai-Man Po, and Kwok-Wai Cheung [6] have introduced a new mechanism known as Dominant Color Structure Descriptor (DCSD) for efficient representation by combining both the color and the spatial structure information with single descriptor. DCSD has combined the compactness of dominant color descriptor and the retrieval accuracy of colour structure descriptor.…”