Development of High Accuracy Classifier for the Speaker Recognition System

Abstract

Speech signal is enriched with plenty of features used for biometrical recognition and other applications like gender and emotional recognition. Channel conditions manifested by background noise and reverberation are the main challenges causing feature shifts in the test and training data. In this paper, a hybrid speaker identification model for consistent speech features and high recognition accuracy is made. Features using Mel frequency spectrum coefficients (MFCC) have been improved by incorporating a pitch frequency coefficient from speech time domain analysis. In order to enhance noise immunity, we proposed a single hidden layer feed-forward neural network (FFNN) tuned by an optimized particle swarm optimization (OPSO) algorithm. The proposed model is tested using 10-fold cross-validation over different levels of Adaptive White Gaussian Noise (AWGN) (0-50 dB). A recognition accuracy of 97.83% was obtained from the proposed model in clean voice environments. However, a noisy channel is realized with lesser impact on the proposed model as compared with other baseline classifiers such as plain-FFNN, random forest (RF), K -nearest neighbour (KNN), and support vector machine (SVM).

Speaker Recognition Assessment in a Continuous System for Speaker Identification

“…The vocal aperture not only influences the geographic roots, but also the voice pitch, the flow and structural characteristics [9]. The speaker recognition systems are also highly used to track a speaker's presence.…”

“…These short-term sensory functions were formulated for the identification of the signal community and the emotion of certain speaker functions. Work was done with entropy and probabilistic methods for recognizing wrath [9]. In order to produce function score and calculate relative weights the author has processed the acoustic and linguistic features.…”

Enhancement and modification of automatic speaker verification by utilizing hidden Markov model

“…A person's claimed identity may be verified or authenticated using recognition in either identification or identification mode (the system determines the best match from the full enrolled population) are examples of this [6], [7]. Since the 1970s, techniques of speaker recognition have been deemed "classic" in the field of biometrics [8], [9]. The acoustic characteristics of each speaker are extracted by speaker identification systems from the spoken stream [10]- 1398 [12].…”

“…The acoustic characteristics of each speaker are extracted by speaker identification systems from the spoken stream [10]- 1398 [12]. The following characteristics are reflected in: i) Anatomy: the geometrical and size forms of the voice lips, velum, lungs, teeth, tongue and lips cords [13]; ii) learned behavioral patterns: the manner in which one speaks [14]; Learning: the manner in which one speaks [9].…”

Retracted: Development of High Accuracy Classifier for the Speaker Recognition System

“…This article has been retracted by Hindawi following an investigation undertaken by the publisher [ 1 ]. This investigation has uncovered evidence of one or more of the following indicators of systematic manipulation of the publication process: Discrepancies in scope Discrepancies in the description of the research reported Discrepancies between the availability of data and the research described Inappropriate citations Incoherent, meaningless and/or irrelevant content included in the article Peer-review manipulation …”