Search

Relative fundamental frequency (RFF) is an acoustic measure that captures the instantaneous changes in fundame nt al frequency (F0) a s a speaker transitions from voicing into and out of [...], Purpose: The effect of phonetic context on relative fundamental frequency (RFF) was examined, in order to develop stimuli sets with minimal within-speaker variability that can be implemented in future clinical protocols. Method: Sixteen speakers with healthy voices produced RFF stimuli. Uniform utterances consisted of 3 repetitions of the same voiced sonorant-voiceless consonant-voiced sonorant speech sequence; moderately variable sentences contained speech sequences with a single voiceless phoneme (/f/, /s/, /∫/, /p/, /t/, or/k/); highly variable sentences were loaded with speech sequences using multiple phonemes. Effects of stimulus type (uniform, moderately variable, and highly variable) and phoneme identity (/f/, /s/, /∫/, /p/, /t/, and /k/) on RFF means and standard deviations were determined. Results: Stimulus type and the interaction of vocal cycle and stimulus type were significant for RFF means and standard deviations but with small effect sizes. Phoneme identity and the interaction of vocal cycle and phoneme identity on RFF means and standard deviations were also significant with small to medium effect sizes. Conclusions: For speakers with healthy voices, uniform utterances with /f/ and /∫/ have the lowest standard deviations and thus are recommended for RFF-based assessments. Future work is necessary to extend these findings to disordered voices. Key Words: voice, speech production, assessment, acoustics

Coarticulation data for Catalan reveal that, while being less sensitive to vowel effects at the consonant period, the alveolar trill [r] exerts more prominent effects than [l] on both adjacent [...]

The ability to quantify voice quality precisely and quickly is critical for a number of applications, such as for the clinical assessment of voice disorders as well as automatic detection [...], Purpose: In this experiment, a single comparison stimulus was developed as a reference in a perceptual matching task for the quantification of breathy voice quality. Perceptual judgments of a set of synthetic voice samples were compared to previous data obtained using multiple comparison stimuli 'customized' for different voices (Patel, Shrivastav, & Eddins, 2010). Method: Five male and 5 female samples of the vowel /a/ were selected from the Kay Elemetrics Disordered Voice Database and resynthesized using a Klatt synthesizer. Eleven samples were created for each base voice by manipulating the aspiration noise level. Five samples from each continuum were evaluated in a perceptual matching task in which a single sawtooth and noise comparison stimulus was used to obtain breathiness judgments. Linear regression was used to compare measurements obtained using the new comparison stimulus against the customized comparison stimuli. Results: Results indicated that the noncustomized sawtooth comparison provides reliability and perceptual distances between stimuli similar to those obtained using customized comparison stimuli. Conclusion: A single-variable matching task using a single comparison stimulus can be used to obtain perceptual estimates of breathiness across voices and experiments in a laboratory setting. This technique will help develop models of voice-quality perception. Key Words: voice quality, matching, breathy, single-variable

Stop consonants are produced by temporarily occluding the vocal tract at a specific location along its length. The ability of a talker to grade the shape, degree, and timing of [...], Purpose: The present study was designed to investigate the relation of formant transitions to place-of-articulation for stop consonants. A speech production model was used to generate simulated utterances containing voiced stop consonants, and a perceptual experiment was performed to test their identification by listeners. Method: Based on a model of the vocal tract shape, a theoretical basis for reducing highly variable formant transitions to more invariant formant deflection patterns as a function of constriction location was proposed. A speech production model was used to simulate vowel-consonant-vowel (VCV) utterances for 3 underlying vowel-vowel contexts and for which the constriction location was incrementally moved from the lips toward the velar part of the vocal tract. These simulated VCVs were presented to listeners who were asked to identify the consonant. Results: Listener responses indicated that phonetic boundaries were well aligned with points along the vocal tract length where there was a shift in the deflection polarity of either the 2nd or 3rd formant. Conclusions: This study demonstrated that regions of the vocal tract exist that, when constricted, shift the formant frequencies in a predictable direction. Based on a perceptual experiment, the boundaries of these acoustically defined regions were shown to coincide with phonetic categories for stop consonants. KEY WORDS: vocal tract shape, consonant identification, formant transitions

Purpose: Age-related changes in tongue function may contribute to dysphagia in elderly people. The authors' purpose was to investigate whether aged rats that have undergone tongue exercise would manifest increased protrusive tongue forces and increased genioglossus (GG) muscle fiber cross-sectional areas. Method: Forty-eight young adult, middle-aged, and old Fischer 344/Brown Norway rats received 8 weeks of tongue exercise. Protrusive tongue forces were measured before and after exercise. GG muscle fiber cross-sectional area was measured in exercised rats and was compared with cross-sectional areas in a no-exercise control group. Results: A significant increase in maximum tongue force was found following exercise in all age groups. In addition, a trend for increased GG muscle fiber cross-sectional area and a significant increase in variability of GG muscle fiber cross-sectional area was identified postexercise. Conclusion: The findings of this study have implications for treatment of elderly persons with dysphagia using tongue exercise programs. Specifically, increases in tongue force that occur following 8 weeks of progressive resistance tongue exercise may be accompanied by alterations in tongue muscle fiber morphology. These changes may provide greater strength and endurance for goal-oriented actions associated with the oropharyngeal swallow and should be investigated in future research. KEY WORDS: tongue, aging, rat, tongue exercise, genioglossus, swallowing, Deglutition, Sarcopenia, the age-related loss of skeletal muscle mass, strength, and function (Rosenberg, 1997) is caused by multiple factors (Doherty, 2003). A major component of sarcopenia is muscle atrophy, which is [...]

Purpose: This article integrates published acoustic data on the development of vowel production. Age specific data on formant frequencies are considered in the light of information on the development of [...]

Abstract Language serves as a primary means of human adaptation. This paper critically examines attempts to identify certain phonological aspects of language systems as particularly adaptive. These postulate that the [...]

Brain mechanisms in humans group together acoustical frequency components both in the spectral and temporal domain, which leads to the perception of auditory objects and of streams of sound events that are of biological and communicative significance. At the perceptual level, behavioral data on mammals that clearly support the presence of common concepts for processing species-specific communication sounds are unavailable. Here, we synthesize 17 models of mouse pup wriggling calls, present them in sequences of four calls to the pups' mothers in a natural communication situation, and record the maternal response behavior. We show that the biological significance of a call sequence depends on grouping together three predominant frequency components (formants) to an acoustic object within a critical time window of about 30-ms lead or lag time of the first formant. Longer lead or lag times significantly reduce the maternal responsiveness. Central inhibition seems to be responsible for setting this time window, which is also found in numerous perceptual studies in humans. Further, a minimum of 100-ms simultaneous presence of the three formants is necessary for occurrence of response behavior. As in humans, onset-time asynchronies of formants and formant durations interact nonlinearly to influence the adequate perception of a stream of sounds. Together, these data point to common rules for time-critical spectral integration, perception of acoustical objects, and auditory streaming (perception of an acoustical Gestalt) in mice and humans. auditory objects | auditory streaming | formant grouping | time-critical processing | sound communication

Prevocalic and postvocalic /l/ were investigated in three adult subjects utilizing a combination of electropalatographic and acoustic techniques. Results indicated that prevocalic /l/ was characterized by both alveolar and lateral lingua-palatal contact, while postvocalic /l/ was primarily alveolar contact only. Acoustically, prevocalic/l/had a lower first formant and higher second formant than postvocalic /l/. In addition, the second and third formants were often weak or absent for prevocalic but not postvocalic /l/. Vowel context had a greater effect on the electropalatographic and acoustic characteristics of prevocalic than postvocalic /l/. Models that relate physiological and acoustical aspects of speech were utilized to account for the observed results.

Inferences were made regarding vocal tract vowel space during fluently produced utterances through examination of the first two formant frequencies. Fifteen adult males served as subjects, representing separate groups of untreated and treated individuals who stutter and nonstuttering controls. The steady-state portion of formant one (F1) and formant two (F2) was examined in the production of various CVC tokens containing the vowels /i/, /u/, and /a/. Vocal tract vowel space was estimated three ways. The first analysis scheme involved measurement of formant frequency spacing. The second measure involved calculating the area of the vowel space triangle. The third measure was based on calculating the average Euclidean distance from each subject's midpoint "centroid" vocal tract position to the corresponding /i/, /u/, and /a/ points on the vowel triangle. The formant frequency spacing measures proved to be most revealing of group differences, with the untreated stutterers showing significantly greater vowel centralization than the treated group and control group. Discussion focuses on the vocal tract articulation characterizing fluent speech productions and possible treatment implications for persons who stutter. KEY WORDS: stuttering, fluency, acoustics, formant frequency, vowel centralization

Vowel duration in open and closed syllables was investigated in groups of ten normal 5-, 8-, and 10-year-old children, and adults. Productions of three syllables (bee, bead, beet), analyzed acoustically for vowel duration, indicated that for all age groups, mean vowel durations for 'bee' and 'bead' were significantly different from 'beet' but were not different from each other. Developmental effects were noted in mean and variability of vowel duration. In addition, a context effect was observed, with adult-like values for both mean and variability obtained at an earlier age for 'beet' than for 'bee' and 'bead.' These results suggest that both phonemic and phonetic factors are important in accounting for developmental changes in mean and variability of vowel duration.

This paper presents a novel analog scheme suitable for the real-time estimation of formant frequencies. Formant tracking is based on a feedback technique which uses both the amplitude and phase characteristics of two stagger-tuned bandpass filters to give an improved dynamic behavior. The implementation of the system requires a small number of components, and is practical for low-power applications. An analysis of the static and dynamic behavior is given for sinusoidal input signals. The transient response is independent of the amplitude level of the input signal. The system is designed for second formant detection in a cochlear prosthesis system.

This paper presents the second part of the analysis of a feedback control system for real-time formant estimation. The system behavior is analyzed for an input signal composed of two sinusoids. If the frequency difference between the two input spectral lines is sufficiently great and the amplitude ratio is within certain limits, a hysteresis effect occurs. Then the system shows a tendency to select one of the two input spectral lines. The existence of the second line has only little influence on the accuracy of the detection of the selected line. From the analysis, conclusions of the system behavior regarding formant estimation can be drawn. A design example for second formant detection is simulated and compared with the results obtained by simulation of a zero-crossing system for F2 estimation and LPC analysis.

Nine subjects of a study were requested to repeat or to create verbs related to nouns read by a researcher while magnetic resonance images of the left inferior frontal lobe were developed. The performance of tasks was accompanied by an increase in signal with much higher activation for verb generation than for repetition. Large activation was observed for a homologous area in the right frontal cortex but not for control areas in the calcarine complex.

A method for speaker normalization and adaption using connectionist networks is developed. A speaker-specific linear transformation of observations of the speech signal is computed using second-order network units. Classification is accomplished by a multilayer feedforward network that operates on the normalized speech data. The network is adapted for a new talker by modifying the transformation parameters while leaving the classifier fixed. This is accomplished by back-propagating classification error through the classifier to the second-order transformation units. This method was evaluated for the classification of ten vowels for 76 speakers using the first two format values of the Peterson-Barney data. A classifier optimized on unnormalized data led to a recognition accuracy of 78.3%. When adapted to each speaker, the accuracy improved to 95.3%. Another classifier, optimized on linearly normalized data, resulted in a recognition accuracy of 93.2%. When adapted for each speaker from initial transformation parameters estimated by various methods, the accuracy improved to 96.6%. When the speaker-dependent transformation and nonlinear classifier were simultaneously optimized, a vowel recognition accuracy of as high as 97.5% was obtained. The results suggest that rapid speaker adaptation resulting in high classification accuracy can be accomplished by this method.

In a previous radiographic study of 4 deaf women, some aberrant features in vocal tract configuration were identified for vowels produced with excessive pharyngeal resonance. These features included neutralization of tongue position, elevation of the hyoid, and a retraction of the tongue, associated with a deflection of the epiglottis in the lower pharynx. The vowels, produced simultaneously with X-ray exposure, were analyzed spectrographically to study acoustic correlates of the vocal tract deformation. Comparisons of the formants for vowels /i/ /u/ and /a/ produced by the deaf women with mean formant values for these vowels produced by normal-hearing women revealed no consistent pattern of second-formant deviation. Formant structure evaluated on isovowel lines disclosed consistent neutralization of vowels, with [F.sub.2] values clustering in the 1500-2100 Hz frequency range, which is attributed to the observed restricted horizontal movements of the tongue within the oral and pharyngeal cavities. If such restrictions affect the production of all vowels, a lower [F.sub.2] might be assumed for the front vowels, which normally have a high [F.sub.2]; a higher [F.sub.2] frequency would be anticipated for back vowels, which normally have a low [F.sub.2] The limited sample studied supports this assumption.

Previous literature indicates that a difference may exist between format frequencies ([F.sub.1] and [F.sub.2]) of children, young adult, and elderly speakers. The purpose of this study was to compare [F.sub.1] and [F.sub.2] of 3 young, 6 young adult, and 3 elderly female speakers for the /i/, /ae/, /u/, and /a/ vowels. Analysis indicates a trend towards vowel reduction across the life span. These findings support previous research regarding age-associated acoustic changes as well as support for the possible anatomical and physiological alterations which may influence such changes.

This study was designed to investigate the effects of vocal [f.sub.o] on 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 [alpha] 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. KEY WORDS: spectral noise level, SNL, vocal roughness, pitch

Spectra of vowels were analyzed to determine whether differences exist between the spectral slope of vowels produced by tracheoesophageal (TE) speakers and normal speakers and, if so, to quantify such differences. The linear predictive autocorrelation method was used to calculate smoothed spectra and the spectra were normalized with respect to a low frequency component. Comparisons between normalized spectral energy within a selected high frequency range revealed that energy within this frequency range for vowels produced by TE speakers was significantly higher than that produced by normal speakers. A least-square distance matching procedure was used to quantify speaker group differences in the spectral slope of vowels. Average spectra of vowels produced by the normal speakers could be matched to average spectra of vowels produced by the TE speakers by decreasing the spectral slope of their vowels by 2-3 dB/octave.

Among the distinctive features for consonants, we distinguish a set of primary features that are perceptually the most salient. The strength with which a primary feature is manifested in a given sound is influenced by the secondary features that co-occur with it. The features [sonorant], [continuant] and [coronal] are designated as primary, and this designation applies in all of the eight possible configurations in which they occur. The features [anterior] and [lateral] are also designated as primary, but only under restricted conditions. We examine systematically which combinations of secondary features provide maximal enhancement of the acoustic manifestations of the primary features. This analysis leads to an inventory of eleven preferred feature combinations or segments. There is a close fit between this preferred inventory and a list of the most frequently occurring segments in the languages of the world, based on Maddieson 1984. This match constitutes support for the view that the frequently occurring feature combinations in the languages of the world come about because those combinations maximize perceptual distinctiveness through the mechanism of feature enhancement. It is suggested that these concepts of enhancement and of preferred feature combinations are directly relevant to the notion of markedness.

Voice is one of the most complex structural and functional areas in language. A study was conducted to analyze two passive voice sub-varieties in Ukrainian, namely, the -no/-to passive and the agreeing passive. The study analyzed the -no/-to Ukrainian passive as an actional or dynamic passive and the Ukrainian agreeing passive as a statal or resultive passive. The analysis was restricted to the expression of voice in the perfective aspect. Results are discussed.

The orderly output constraint (OOC) is extraneous. Talkers 'speak in lines' in its absence. Further, there is no perceptual motivation for an OOC; perceivers ignore the linearity between F2 at consonant-vowel onset and F2 in the vowel. In any case, the analogy with bat and barn owl localization systems underlying the theory is extreme, Sussman et al.'s comments to the contrary notwithstanding.

Although the relations between second format (F2) onset and F2 vowel are extremely regular and contain important information about place of articulation of the voiced stops, they are not sufficient for its identification. Using quadratic discriminant analysis of a new data set, it is shown that F3 onset and F3 vowel can also contribute substantial additional information to help identify the consonants.

Sussman et al. describe an ecological property of the speech signal that is putatively functional in perception. An important issue, however, is whether their putative cue is an emerging feature or whether the second formant (F2) onset and the F2 vowel actually provide independent cues to perceptual categorization. Regardless of the outcome of this issue, an important goal of speech research is to understand how multiple cues are evaluated and integrated to achieve categorization.