Kristin Precoda

Photoglottography: A clinical synopsis

Summary Although photoglottography (PGG) has been used as a measure proportional to glottal area, it has not been widely applied in the clinic to study dysphonic patient populations. Historically, PGG has required the insertion of either a light source or photosensor at least to the level of the oropharynx or nasopharynx. This invasive nature of PGG has probably limited its appeal to those who are unwilling to risk injury or discomfort to subjects. Additionally, the effort and time necessary to carefully hand-mark glottal events for analysis has limited its clinical use. This report presents a brief overview of PGG and describes two techniques to help enhance its clinical application: a minimally invasive transoral technique of illumination and an automated technique to identify glottal events. In addition, two possible factors that may confound the interpretation of transoral PGG results were evaluated: the effects of transoral versus transnasal light sources, and effects of change in articulatory configuration on PGG results.

Source characteristics of diplophonia

Diplophonia describes the perception of more than one fundamental frequency component in a voice. Simultaneous photoglottographic (PGG), electroglottographic (EGG), and acoustic signals were collected from five diplophonic subjects. Results indicate that diplophonia is similar to the beat frequency phenomenon caused by independent vibration of the two vocal folds at slightly different frequencies. In the time domain, modeling vocal fold displacement from midline by summing two sine waves produced area waveforms closely resembling the PGG waveforms obtained from the diplophonic subjects. In the frequency domain, the diplophonic voice spectrum bears little resemblance to that of normal phonation. Instead of the normal set of harmonic peaks at integer multiples of the fundamental, the diplophonic spectrum shows a wealth of very closely spaced energy peaks of which the two fundamentals are often the most prominent but rarely the lowest frequency. The spectral envelope of the glottal area waveform may appear t...

Perception of supraperiodic voices

The study of voice quality is built largely upon the assumption of vocal (quasi‐)periodicity. Supraperiodic phonation types challenge traditional concepts of vocal periodicity, and thus are of theoretic and practical importance, especially when measuring or describing pathologic vocal function. Unfortunately, the literature describing such phonation is confusing. It has been argued [Gerratt et al., J. Acoust. Soc. Am. Suppl. 1 83, S66 (1988)] that the traditional term ‘‘diplophonia’’ has been applied to two distinct phonation types. In one (true diplophonia), the two vocal folds vibrate at different rates, producing a waveform resembling a high‐frequency wave modulated by a lower frequency envelope. In the second phonation type (‘‘bicyclicity’’), a pattern of two cycles repeats; cycles within the pattern differ in period and/or amplitude, giving a big–small–big–small appearance to the waveform. To investigate the perceptual characteristics of these phonation types, sets of male and female voices were cons...

Variability in human judgments of foreign accent strength

Speech recognition‐based measures of foreign accent strength have been shown to correlate well with human judgments. However, some automatic measures vary noticeably, while the variability of human judgments is generally unknown. This study examined variability in human ratings of foreign accent strength within speakers. Seventy‐six Japanese adults with varying English pronunciation skills read an average of 65 sentences. Eleven native speakers of American English rated the utterances for foreign accent strength using a five‐point anchored scale. Listeners were selected partly for their abilities both to use the full scale and to give consistent repeated ratings. To measure categorical variability, the entropy of each speaker’s ratings by each listener was calculated. The mean equivalent number of equally likely categories was 2.6; entropy and the speaker’s average rating were strongly related. This could be due to speaker characteristics (poor speakers may vary more) or to properties of the scale (the ex...

Reliability of listeners judging compressed audio

Listener judgments of audio quality are important in designing minimally distorting compression algorithms, yet the reliability of these judgments is often considered questionable. This experiment examines reliability in detail. Nine listeners heard all possible pairs of 15 signals, the original and 14 generated by compression algorithms, for each of two music samples. They rated the similarity of members of a pair on a seven‐point scale—generalizing the idea that high‐quality compression produces output highly similar to the original. Mean intralistener agreement within one scale value on repeated ratings was 85.9% (range 66.7%–100%). Interlistener reliability was examined without one version determined to be a perceptual outlier. The average correlation between listeners’ similarity judgments was 0.57 (range 0.28–0.76). However, multidimensional scaling analyses of judgments of one music sample showed listeners attended to very much the same acoustic characteristics, but weighted them differently. When ...