Voice Pathology Detection Based eon Short-Term Jitter Estimations in Running Speech

Abstract

In this paper, we investigate the use of jitter estimation over short time intervals (short-term jitter) for voice pathology detection in the case of running or continuous speech. Short-term jitter estimations are provided by the spectral jitter estimator (SJE), which is based on a mathematical description of the jitter phenomenon. The SJE has been shown to be robust against errors in pitch period estimations, which makes it a good candidate for measuring jitter in continuous speech. On two large databases of sustained vowel recordings from healthy and pathological voices, we suggest a threshold for the SJE for pathology detection based on cross-database validation. Applying that to a database of continuous speech (reading text) from normophonic and dysphonic speakers, a second threshold and new features are suggested for monitoring jitter in continuous speech. Detection performance of the suggested thresholds and features was evaluated using receiver operating characteristic curves and their discriminative efficiency between healthy and pathological voices was judged using the area under the curve index. In terms of area under the curve, the suggested features for reading text provide a discrimination score of about 95%, while the second threshold provides a classification rate of 87.8%. Furthermore, estimated short-term jitter values from reading text were found to confirm the studies showing a decrease of jitter with increasing fundamental frequencies, and the more frequent presence of high jitter values in the case of pathological voices as time increases.

Glottal source processing: From analysis to applications

“…Results showed that there were significant differences in the performance of the various algorithms, the best techniques being not necessarily the most commonly used. In (Vasilakis and Stylianou (2009)), the use of the spectral short-term jitter estimator is proposed to discriminate voice pathologies in running or continuous speech, leading to an interesting performance. Jitter values were found to confirm studies showing a decrease of jitter with increasing fundamental frequencies, and the more frequent presence of high jitter values in the case of pathological voices as time increases.…”

Feature enhancement for classifier optimization and dimensionality reduction

“…SVM classifier with MFCC and modulation spectra based features giving accuracy of 81.70% is reported [17]. Spectral Jitter estimator as local aperiodicity measure is studied in [18] giving AUC (area under curve) as 84.65%. In this work number of frames over and below the threshold were measured and thresholds for the "Over" feature were recognized as the one that results in high discrimination for normal versus pathological voices.…”

Improved Algorithm for Pathological and Normal Voices Identification

“…The first approach consist to compare acoustics parameters between normal and abnormal voices such as fundamental frequency, jitter, shimmer, harmonic to noise ratio, intensity [3]. The second approach is IJECE ISSN: 2088-8708  Improved Algorithm for Pathological and Normal Voices Identification (Brahim Sabir) 239 the parametric and non-parametric for features selection [4], [5].…”