Pathological Voice Assessment

Abstract

While there are number of guidelines and methods used in practice, there is no standard universally agreed upon system for assessment of pathological voices. Pathological voices are primarily labeled based on the perceptual judgments of specialists, a process that may result in different label(s) being assigned to a given voice sample. This paper focuses on the recognition of five specific pathologies. The main goal is to compare two different classification methods. The first method considers single label classification by assigning a new label (single label) to the ensembles to which they most likely belong. The second method employs all labels originally assigned to the voice samples. Our results show that the pathological voice assessment performance in the second method is improved with respect to the first method.

Tabby Talks: An automated tool for the assessment of childhood apraxia of speech

“…Jitter (perturbations in pitch) and shimmer (perturbations in amplitude) are used to assess the severity of voice pathologies (Gelzinis et al, 2008;Manfredi et al, 2000). Good results have been obtained using the differential Teager energy operator (Hansen et al, 1998) and Mel frequency cepstral coefficients (Dibazar et al, 2006;Godino-Llorente and Gomez-Vilda, 2004) as well as time-frequency decompositions (Umapathy et al, 2005) and nonlinear dynamics (Henriquez et al, 2009). The autocorrelation method was found to perform the best in tracking pitch perturbations in different pathological voices as it resulted in the least amount of errors (Seung-Jin et al, 2007).…”

Voice Pathology Recognition and Classification using Noise Related Features

“…Based on the human knowledge of the sounds, MFCC does a frequency analysis of the signal. By listening to the signal an experienced therapist can detect the presence of a speech disorder [2]. For each frame, the extraction procedure is done after a 16 kHz interpolation, with a bank of 29 Mel filters and a 25 ms with a 10 ms step, to get 12 MFCC plus log-energy, Delta and Delta seconds.…”

“…A normal binary / pathological classification of vocal samples [1,2] has been proposed in the literature, the best performances are obtained by using specific parameters of the HMM classification. However, few studies that have classified the pathologies [3] and the obtained results were not effective.…”

Voice pathology distinction using autoassociative neural networks

“…Most of the methods in the literature focus on binary classification to detect whether the voice is classified as normal or pathological, they do not detect the type of voice pathology. The differentation between 5 voice pathologies is presented in the paper [17]. The classification accuracy is in the range 61% and 69%.…”