Search

Direct magnitude estimation (DME) and equal-appearing interval (EAI) ratings of the perceived roughness of isolated sustained vowel samples (/a/ and /i/) were obtained. The linear and curvilinear relationship between the resulting EAI and DME ratings was then examined. That relationship was found to be more strongly curvilinear than linear, a characteristic of a prothetic continuum. The amount of variance accounted for by a curvilinear model of the relationship, however, was minimally more than that accounted for by a linear model. The findings thus suggest that the roughness continuum manifests no more than "a little" protheticness, if any. It does not seem clear from this initial study that there is greater validity associated with roughness ratings obtained by the DME than the EAI method. Rather, the findings suggest that a small degree of such advantage may exist, but it may be too small to be of practical consequence in many instances.

This study was designed to investigate the formant frequencies of phonated and whispered productions of five test vowels (/i/,/u/, /æ/, /Δ/, and //). Each test vowel was sustained twice in isolation—once phonated, once whispered—by 20 adult female subjects. The phoneme represented by each recorded production was identified independently by 11 listeners. Only those samples identified by 6 or more of the listeners as the vowel intended were retained for a further (acoustic) analysis. An acoustic spectrum of each retained sample was obtained to permit formant measurements. To provide the clearest formant delineation possible in our lab, the phonated samples were analyzed by broadband spectrography and the whispered samples by very narrowband spectrography. This report presents the formant frequencies (FI-F3) of the test productions as measured from the acoustic vowel spectra and the formant-freqneney differences between the pbonated and whispered productions. In general, the results showed a trend for whispered vowel formants to be higher in frequency than phonated vowel formants, but that trend was only strongly evident for F1.

This study of children's speech was designed to investigate peak oral-nasal airflow during the production of stop-plosive consonants in isolated syllable contexts. The subjects were 30 normal-speaking children: five girls and five boys at each of three age levels (8, 9, and 10 years). Each subject produced each of six stop-plosive consonants (/p/, /b/, /t/, /d/, /k/, and/g/) in each of three syllable contexts (CV, VCV, and VC) with each of two vowels (/i/ and/a/). A pneumotachometer and an oscillographic recorder were used to obtain permanent airflow records of the test syllables. It was found that the peak flows varied according to the age of subjects, consonant voicing, place of consonant production, consonant position within tire test syllable, and the vowel in combination with the test consonant. Results were compared to related findings previously obtained for children and adults.

This study was designed to investigate the formant frequencies and amplitudes of phonated and whispered productions of five test vowels (/i/, /u/, /æ/, /ɑ/, and /ʌ/). Each test vowel was sustained six times in isolation, three times in whisper and three times in phonation, by 15 adult male subjects. The phoneme represented by each recorded production was identified independently by 11 listeners. An acoustic spectrum of each sample was obtained through the use of a real-time narrow-band spectrum analyzer to permit formant measurements. This report presents the formant frequencies and amplitudes (F1to F3) of the test productions as measured from the acoustic vowel spectra and the formant frequency and amplitude differences between the phonated and whispered productions. In general, the results showed a trend for whispered vowel formants to be higher in frequency than phonated vowel formants, but that trend was only strongly evident for F1.

Two experiments were designed to study whispered and phonated vowel perception. In each experiment, the whispered and phonated test vowel samples were produced by the same subjects. In experiment 1, vowel samples produced by adult female subjects were examined. In experiment 2, vowel samples produced by adult male subjects were examined. In both experiments, whispered and phonated isolated vowels were identified correctly by individual judges in a majority of instances. Phonated vowel samples, generally, were identified correctly more often than whispered vowel samples. The vowels /i, æ, u/ were identified best among phonated vowels; the vowel /i/ was identified best among whispered vowels.

The effects of vocal pitch on the perceived roughess and the acoustic spectral noise levels of two test vowels were investigated. Ten normal-speaking young adult subjects (5 male and 5 female) each produced at one intensity the high vowel /u/ and the low vowel /a/ at four test pitches. The pitches ranged, for each subject, from habitual pitch to a pitch one octave higher. Spectral noise level (SNL) measures and listener ratings of roughness were obtained for each vowel sample, and the fundamental frequency of each sample was measured. It was found that both the mean SNL and the perceived roughness of the test phonations diminished significantly (P< 0·05) when vocal pitch was raised by an octave. This finding held for the two sexes (considered separately) and for both test vowels. When the data were compared across sexes, it was found for each test vowel that the roughness ratings and the SNL means were more closely related to the relative pitch (re subjects’ habitual pitch) at which the test phonations were produced than to the absolute fundamental frequency of phonation.

This study was designed to investigate the effects of vocal foon vowel spectral noise level (SNL) and perceived vowel roughness for subjects in high- and low-pitch voice categories. The subjects were 40 adult singers (10 each sopranos, altos, tenors, and basses). Each produced the vowel /a/ in isolation at a comfortable speaking pitch, and at each of seven assigned pitches spaced at whole-tone intervals over a musical octave within his or her singing pitch range. The eight /a/ productions were repeated by each subject on a second test day. The SNL differences between repeated test samples (different days) were not statistically significant for any subject group. For the vowel samples produced at a comfortable pitch, a relatively large SNL was associated with samples phonated by the subjects of each sex who manifested the relatively low singing pitch range. Regarding the vowel samples produced at the assigned-pitch levels, it was found that both vowel SNL and perceived vowel roughness decreased as test-pitch level was raised over a range of one octave. The relationship between vocal pitch and either vowel roughness or SNL approached linearity for each of the four subject groups.

Direct magnitude estimation (DME) and equal-appearing interval (EAI) ratings of perceived roughness and spectral noise level (SNL) measurements of isolated sustained vowel samples (/a/ and /i/) were obtained. The linear and curvilinear relationships between the resulting roughness ratings and SNL were then examined. In the present study, the comparison between the two models resulted in conflicting evidence regarding the model that best described the relationship between roughness ratings and SNL measurements. In all comparisons, however, a significant linear component was indicated. Additionally, when a significant curvilinear component was indicated, the difference between the lines and curves that were fit to the data was minimal. The present findings suggest, therefore, that a high degree of linearity exists between both DME and EAI roughness ratings and SNL.

This study investigated the relationship between spectral noise level (SNL) and perceived roughness in sustained vowels phonated by children. Twenty male children, ten presenting normal vocal quality and ten presenting vocal roughness associated with benign vocal cord lesions, served as subjects. The children individually sustained each of the vowels /i/, /u/, /∧/, /a/, and /æ/ at one intensity and at two pitch levels (comfortableand high). Each vowel recording was analyzed to produce a 3-Hz bandwidth, amplitude-by-frequency acoustic spectrum. The levels of inharmonic (noise) components were measured for the range 100 to 2600 Hz within each vowel spectrum and the mean of those measures provided an index of vowel spectral noise level. Each recorded sample was also rated for degree of roughness, by 11 trained judges, using a five-point, equal-appearing intervals scale. The fundamental vocal frequency (fo) of each production also was measured. The results indicated that four subjects with vocal pathology phonated their vowel productions with a relatively extreme fowhen compared to subjects with normal voice quality. It was also found that vowel roughness and SNL differed predictably for individual vowels and that, especially for normal-speaking subjects, roughness and SNL decreased with an increase in fo. Further, for both subject groups, SNL measures were correlated positively and moderately highly with perceived roughness.

In this study the effects of vowel roughness on the level of harmonics in test vowel spectra were investigated. Twenty normal-speaking adult males sustained productions of each of the vowels /u/, /i/, /Λ/, /a/, and /æ/ first normally and then with simulated abnormal vocal roughness, at one intensity. A tape recording of each production was rated for roughness on an equal-appearing intervals scale by 11 judges and was also analyzed to obtain its 3-Hz bandwidth amplitude-by-frequency acoustic spectrum. The median roughness rating, the level of the first five spectral harmonics, and the foof each production were obtained. It was found that the level of the first three harmonics for each test vowel diminished with an increase in roughness (as was hypothesized from earlier results), but the higher harmonics of /u/ and /i/ showed some reversals. A moderately large negative correlative relationship was shown between lower-harmonic levels and median vowel roughness. A possible explanation of the higher-harmonic reversals was offered. The results suggest that spectral noise level measurements may provide a more reliable acoustic index of vowel wave aperiodicity and vowel roughness than measurements of harmonic level.

Acoustic wave-period variation (jitter) and acoustic wave-amplitude variation (shimmer) associated with vowel pronations representing a range of vocal roughness were investigated. Twenty normal-speaking adult males phonated each of the vowels /u/, /i/, /Λ/, /a/, and /æ/, first normally and then with simulated abnormal vocal roughness. Twenty hoarse adult males, each presenting a medically diagnosed laryngeal pathology, also produced each of the five test vowels. To provide a measurable presentation of the frequency and amplitude variations of interest, each recorded vowel was band-pass filtered to isolate the fundamental frequency component. Relations of the jitter and shimmer indices (obtained from the filtered vowel waves) to acoustic spectral noise levels and to roughness ratings for the vowel phonations were studied. The findings supported the hypothesis that increases in vowel acoustic wave variability (estimated by period or amplitude variation or both) are associated with increases in vowel spectral noise levels and perceived vowel roughness. The findings also suggested, for most of the vowels studied, that cyclic peak amplitude variation may provide a better index of perceived roughness than cyclic period variation. Vowel spectral noise levels, however, may provide a more clinically useful indicant of vowel roughness than the waveform variability indices derived from the filtering procedure employed in this study.

Twenty normal-speaking adult males sustained productions of each of the vowels /u/, /i/, /Λ/, /a/, and /æ/ first normally and then with simulated vocal roughness at one intensity. A tape recording of each production was rated for roughness on a five-point equal-appearing-intervals scale by 11 trained judges and was also analyzed to produce its 3-Hz bandwidth frequency-by-amplitude acoustic spectrum. The median roughness rating and the level of inharmonic spectral components, i.e., noise components, for each production were examined and related. Noise levels for the productions of each vowel averaged over selected spectral ranges between 100 Hz and 8000 Hz correlated highly with the median roughness ratings for those productions. Multiple correlation coefficients indicating the relationship between the median roughness rating and multiple measures of spectral noise in the range from 100 Hz to 2600 Hz for the individual productions of each test vowel were high (≥0.97) and significant for all five vowels.

Twenty normal-speaking adult females sustained seven-second productions of the vowels /u/, /i/, /Λ/, /a/, and /æ/ first normally and then with simulated vocal roughness at one intensity. Recordings of the vowels were rated for roughness on a five-point equal-appearing intervals scale by 11 trained judges. Each production was also analyzed to produce a 3-Hz bandwidth frequency-by-amplitude acoustic spectrum in which the level of inharmonic, i.e., noise, components was measured in dB SPL. Noise levels, averaged over selected frequency ranges from 100 to 8000 Hz, for each vowel production correlated highly with the median roughness ratings for the productions. High (≥0.97) and significant multiple correlation coefficients were obtained between each vowel’s 100-Hz section noise levels from 100 to 2600 Hz and its median roughness rating.