Consideration of the relationship between the fundamental frequency of phonation and vocal jitter.
Explore the source record for details and available documents.
Biomedical subjects
Publications and source records attributed to R J Baken.
Explore the source record for details and available documents.
Signal-averaging and autocorrelation analysis revealed that the cardiovascular system exerts a modest but consistent influence on vocal fundamental frequency (Fo), accounting for approximately 0.5% to 20% of the absolute Fo perturbation (jitter) measured during a sustained phonation. There was also a marked trend for this percentage to decrease with increasing vocal Fo. Estimated mean "deterministic jitter" (Jd) values of 3.7 microsec (SD = 3.2) and 0.9 microsec (SD = 0.5) were derived from 6 normal male and 6 normal female subjects, respectively, with an overall mean of 2.3 microsec (SD = 2.7). These values represent approximately 6.9% of the mean total jitter for men and 2.4% of the mean total jitter for women, or about 4.6% for all subjects. The results are discussed in terms of their significance regarding more reliable vocal jitter measurement.
Signal-averaging techniques reveal that the vocal fundamental frequency (F0) of a sustained vowel is modulated over a period equal to that of the speaker's heart cycle. Average F0 deviation varies in an orderly way from about 0.5% to about 1.0% as F0 changes. Location of the peak deviation in the time frame of the heart cycle also changes systematically with vocal F0. Modulation of the vocal F0 is likely to be caused by pressure-related changes in the stiffness of the vascular bed of the vocal folds and by alterations of the geometry of the thyroarytenoid muscle produced by periodic vascular engorgement.
In addition to cycle-by-cycle (jitter) and long-term (intonation contour and declination) changes, vocal fundamental frequency (Fo) is known to vary during moments of production of individual phones. This study explored the relationship between intra-oral pressure and Fo during the production of the English voiced fricatives (v), (z), (o), and (3). Target words were embedded in a carrier phrase spoken with three different patterns of sentence stress. Fo changed at a mean rate of -3.59 Hz/cmH2O and -7.96 Hz/cmH2O in men and women, respectively. No significant difference was observed among the different fricatives nor among the several stress patterns. A significant sex effect, not observed in a prior related study, was eliminated by conversion of the Fo data to semitones. The observed magnitudes of the ratios of Fo change to pressure change are consistent with several earlier studies that explored the effect of passive transglottal pressure changes on Fo. The present findings imply that, although Fo regulation is involved in the generation of different intonation contours, the laryngeal system is not compensated to maintain Fo in the face of the transitory changes in vocal-tract aerodynamics that accompany voiced fricative production.
The timing of prephonatory movements of the larynx, rib cage, and abdomen was examined in order to gain insight into the contribution of the vocal folds to the posturing of the chest wall. A simple stimulus-response paradigm was used in eliciting brief utterances--/a/ and /ha/--from six adult males. Chest wall movements were observed using mercury strain gages while simultaneous electroglottographic and airflow records provided information about vocal fold behavior. Independence of prephonatory laryngeal and chest wall behavior was demonstrated. Laryngeal adjustment preceded the start of vocal fold oscillation by a constant amount of time, whereas the time of onset of the chest wall adjustment varied as a function of the utterance type. The qualitative characteristics of prephonatory chest wall posturing were unaffected by altering glottal configuration requirements. Rib cage enlargement occurred during postural adjustment while the vocal folds were abducted (in preparation for /h/). This implies that rib cage enlargement during prephonatory chest wall posturing was not a passive response to abdominal compression.
Examination of some acoustic characteristics of sustained pulse register phonation in normal adults confirmed the existence of two distinct waveform patterns. Trains of similar, highly damped waves with relatively uniform spacing predominated, but the "dicrotic" pattern previously described accounted for about 15% of the total sample. Average differences in the periods of continguous cycles (jitter) were extremely large--on the order of 20% of the fundamental period. Jitter did not vary with fundamental frequency according to the pattern documented in modal register and, thus, jitter ratio is not a useful normalizing descriptor of pulse register perturbation.
This study reports airflow-intensity relationships observed in crescendo and decrescendo phonations produced at three fundamental frequency levels by 10 men. The purpose of the investigation was to study the possible influence of lung volume on airflow-intensity relationships. Of the 30 decrescendo phonations, 26 showed statistically significant correlation coefficients and none showed significant negative correlations. The crescendo phonations, however, showed greater inter- and intrasubject variability. Of the 30 crescendo phonations, 16 showed statistically significant positive correlation coefficients and 6 showed significant negative correlations. Because lung inflation varied directly with intensity in the decrescendo task and indirectly in the crescendo task, it was concluded that lung volume does influence the consistency and strength of relationship between airflow and intensity. The findings are discussed in terms of the influence of lung volume on the relative contributions of glottal resistance and expiratory force to the regulation of subglottal pressure.
The possibility that prephonatory chest wall posturing is abnormal in stutterers was explored by observing rib cage and abdominal hemicircumference changes during the interval between the presentation of a stimulus and the production of/alpha/by a group of stutterers (N = 5). It was found that the patterns of chest wall adjustment for phonation were qualitatively identical in the stutterers and in a comparable group of normal men studied previously. There was, however, a significant difference in the way in which lung volume changed during the execution of the chest wall adjustment. This was considered to be indicative of delayed glottal closure among the stutterers rather than representative of a primary ventilatory disturbance.
Evidence suggests that the fine motor coordinations required to maintain the articulatory-laryngeal adjustments represented by measures of voice onset time (VOT) should change as a function of normal aging. This descriptive study obtained measures of VOT in healthy elderly subjects as a basis for extending the model of temporal speech control to the later years. Three groups of 10 subjects each were used. Group I was composed of subjects 25-39 years old. Groups 2 and 3 included subjects aged 65-74, and over 75, respectively. Subjects produced 30 tokens of three stimulus words, beat, pete, bead, following a carrier phrase. Wide-band spectrograms were prepared and measured, and the VOT data analyzed statistically. The following conclusions were drawn with respect to mean group age. The means of the VOTs did not differ significantly across subject groups. Standard deviations in the means did differ significantly, variability increased with age, both within subjects and between groups. The minimal separation of the phonemic boundaries for /b/ and /p/ decreased significantly as a function of aging, with the /p/ side of the distribution showing the greatest change.
Explore the source record for details and available documents.
Movements of the chest wall during the interval between an acoustic stimulus and the subject's vocal response were examined and timed in eight normal males. The reaction-time interval was divisible into two phases, a latency period with duration independent on chest wall status at the time of stimulus and an adjustment period during which the rib cage and abdomen usually moved oppositionally to achieve a prephonatory postural set. The time required for this adjustment varied significantly with lung volume, but was independent of the ventilatory phaze previously in progress.
The literature has noted that speakers often perceive their own speaking pitch levels differently than listeners perceive them. However, little information is available regarding the specific characteristics of such perceptual differences. Speaking pitch level self-perception was explored in a group of 11 young adult males who served both as talkers and listeners. As a talker, each subject judged his own speaking pitch level in the process of speaking (live judgments) and during taped replay (taped judgments). The subjects' self-rankings in these two tasks and the rank order of taped voices as judged by listeners were compared to fundamental frequency rankings for the voices. The results indicated that the subjects judged their own taped voices in the same way that the listeners judged them, and the judgments corresponded to fundamental frequency rankings. During the live judgments, the subjects avoided extreme self-rankings, preferring to rank themselves closer to an average pitch level. The findings may have clinical significance in the remediation of certain voice disorders.
Respiratory activity during crying was measured in a semilongitudianl study of 10 normal infants aged two days to eight months. Temporal aspects of respiration during crying showed marked change with age: athe duration of the expiratory phase steadily increased, while the duration of the inspiratory phase remained remarkable constant, producing an ever-lower I-fraction. A previous study reported no developmental changes in the temporal aspects of the acoustic cry signal from one to seven months of age. Reasons for the divergent findings of the present study arepresented, and it is suggested that respiratory behavior may be a better indicator of developmental changes in the crying act thanthe acoustic signal.
A technique for estimating lung volume change is described. Voltages proportional to rib cage and abdominal hemicircumferences are produced by Whitney-gage transducers in a standard bridge circuit. The weighted sum of these voltages can be adjusted to produce a good estimate of lung volume change. The instrumentation does not obstruct articulatory movements, load the chest wall, or require restraint of the subject. In addition, it is easy and inexpensive to construct.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
Explore the source record for details and available documents.