Estimation of Voice Perturbation Measures Using Signal Processing Algorithms

Abstract

Human voice, which is the most beautiful instrument, is produced by a complicated mechanism. The quality of the voice is assessed by various techniques. Using voice analysis tools thereby providing an objective assessment of voice. These methods support the clinicians in the diagnosis of laryngeal pathologies associated with the auditory perception and also in deciding the treatment in patients. This paper aims at extraction of voice parameters from voice signals such as Fundamental Frequency (F0 Mean and Standard Deviation) Pitch Perturbation Quotient (Jiiter PPQ), Relative Average Perturbation (Jitter RAP), Amplitude Perturbation Quotient (Shimmer APQ), Pitch Variation (vF0) and Amplitude Variation (vAm) using a novel algorithm leading to an application named Ephphatha was compared with the existing software CSL (Computerized Speech Lab) using t-test analysis. The Bland-Altman plot was done showing the correlation and the limits of agreement between the measures. The outcomes got illustrate that there is no significant difference (p-value>0.05) in F0 and its variation measures. Amplitude and pitch perturbation measures are moderately (>0.01 p-value <0.05) to weakly (p-value<0.01) correlated between the programs. Thus Ephphatha would undoubtedly equip the speech pathologists and therapists with a non-invasive approach to confirm their perceptual observations.

Design of Automatic Scoring System for Oral English Test Based on Sequence Matching and Big Data Analysis

“…e research on automatic oral scoring had achieved remarkable results [8]. In the traditional methods, the design of the automatic scoring system for oral English computer test mainly included the automatic scoring method for oral English computer test based on spectrum analysis and the automatic scoring method for oral English computer test based on wavelet analysis [9].…”