Predicting Intelligibility Deficit in Dysphonic Speech with Cepstral Peak Prominence

Abstract

Objectives: The aim of this study was to examine the potential of cepstral peak prominence (CPP) for predicting the intelligibility deficit in dysphonic speech.

Methods: Sentences from Hearing-in-Noise Test were recorded from 18 speakers with dysphonia and 18 speakers with normal voice. These samples were presented to 60 adults with normal hearing in quiet and noise at signal to noise ratio of +0 dB. Intelligibility was measured by orthographic transcription. Cepstral peak prominence was measured for all samples. Correlation between CPP and intelligibility score was examined.

Results: Intelligibility was significantly lower in dysphonic speech than normal speech in the presence of background noise. The correlation between CPP and intelligibility score was moderate when the intelligibility scores were averaged per speaker.

Conclusions: Cepstral peak prominence only moderately predicts intelligibility deficit in dysphonic speech. Accordingly, CPP alone is not sufficient for describing the deficit.

Keywords: acoustic analysis; background noise; cepstral peak prominence; dysphonia; intelligibility.

Figure 1.

Cepstral peak prominence of (left) normal and (right) breathy voice.

Figure 2.

Correlation plot of cepstral peak prominence and perceptual ratings of dysphonic voice quality.

Figure 3.

Scatter plots showing the relationship between cepstral peak prominence (CPP) and intelligibility scores in (left) noisy and (right) quiet listening conditions. The open circles show averaged scores at the sentence level. The score for each sentence is an average of 15 listeners. The black circles show averaged scores at speaker and listener levels. The scores shown here are averages of 10 sentences and 15 listeners for each speaker. The horizontal error bars show the standard deviations of CPP. The vertical error bars show the standard deviations of intelligibility score.