Vowel-like Stimuli Research Articles

Auditory Spectral Integration in the Perception of Static Vowels

To evaluate potential contributions of broadband spectral integration in the perception of static vowels. Specifically, can the auditory system infer formant frequency information from changes in the intensity weighting across harmonics when the formant itself is missing? Does this type of integration produce the same results in the lower (first formant [F1]) and higher (second formant [F2]) regions? Does the spacing between the spectral components affect a listener's ability to integrate the acoustic cues? Twenty young listeners with normal hearing identified synthesized vowel-like stimuli created for adjustments in the F1 region (/Λ/-/α/, /i/-/ε/) and in the F2 region (/Λ/-/æ/). There were 2 types of stimuli: (a) 2-formant tokens and (b) tokens in which 1 formant was removed and 2 pairs of sine waves were inserted below and above the missing formant; the intensities of these harmonics were modified to cause variations in their spectral center of gravity (COG). The COG effects were tested over a wide range of frequencies. Obtained patterns were consistent with calculated changes to the spectral COG, in both the F1 and F2 regions. The spacing of the sine waves did not affect listeners' responses. The auditory system may perform broadband integration as a type of auditory wideband spectral analysis.

The Effect of Temporal Gap Identification on Speech Perception by Users of Cochlear Implants

This study examined the ability of listeners using cochlear implants (CIs) and listeners with normal hearing (NH) to identify silent gaps of different duration and the relation of this ability to speech understanding in CI users. Sixteen NH adults and 11 postlingually deafened adults with CIs identified synthetic vowel-like stimuli that were either continuous or contained an intervening silent gap ranging from 15 ms to 90 ms. Cumulative d', an index of discriminability, was calculated for each participant. Consonant and consonant-nucleus-consonant (CNC) word identification tasks were administered to the CI group. Overall, the ability to identify stimuli with gaps of different duration was better for the NH group than for the CI group. Seven CI users had cumulative d' scores that were no higher than those of any NH listener, and their CNC word scores ranged from 0% to 30%. The other 4 CI users had cumulative d' scores within the range of the NH group, and their CNC word scores ranged from 46% to 68%. For the CI group, cumulative d' scores were significantly correlated with their speech testing scores. The ability to identify silent gap duration may help explain individual differences in speech perception by CI users.

Alteration of Adults' Subjective Feeling of Familiarity toward Infants' Sounds

Many adults may have lower subjective feelings of familiarity toward infants' vocalizations since infants' sounds are different from those of adults. However, mothers frequently exposed to infants' vocalizations may be more familiar and less averse. To test this hypothesis, 21 mothers (M age = 31.1 yr., SD = 4.3) of infants (M age = 8.2 mo., SD = 3.5), 18 mothers (M age = 34.4 yr., SD = 4.8) of children between two and five years of age (M age = 2.8 yr., SD = 1.0), and 17 women (M age = 29.2 yr., SD = ll.1) with no children were exposed to 20 types of sounds. Of these sounds, 14 were produced by infants. Although the mothers of infants did not recognize sounds as those of an infant's vocalization, they showed higher subjective feelings of familiarity toward the timbres of the vowel-like stimuli than did the other groups. By contrast, the subjective feelings of familiarity for nonspeech sounds did not differ among groups. Maternal experiences may change women's recognition of perceived sounds.

Pitch discrimination in children using cochlear implants

Research in pitch perception by cochlear implant users has focused primarily on postlingually deafened adults. Little is currently known regarding the pitch perception abilities of those who are implanted at a very early age, yet many children receive cochlear implants when they are less than 2 years old. Given the effects of brain plasticity, it is conceivable that pitch perception by early cochlear implant users may be much better than that of their older counterparts. This study examines pitch discrimination thresholds in young cochlear‐implant users between the ages of 4 and 16. A computer game was developed to determine difference limens for fundamental frequency in vowellike stimuli at three referent frequencies: 100, 200, and 400 Hz. Two different tasks were employed: pitchdiscrimination, in which subjects were asked whether two stimuli are same or different, and pitch ranking, in which participants determine which of two stimuli is higher in pitch. Normal‐hearing matches were tested to control for maturational effects, and adult cochlear implant users were also tested to compare findings with other studies. [Work supported by the Nova Scotia Health Research Foundation.]

Spatiotemporal coding based on the responses of individual auditory-nerve fibers

Spatiotemporal coding has been suggested to provide a robust representation of vowel spectra, pitch, and tones in noise. However, much of the quantitative support has come from modeling studies because of the inherent variability in physiological population studies. A spectro-temporal manipulation procedure (STMP) was developed to simulate spatiotemporal patterns from the responses of individual auditory-nerve (AN) fibers to frequency-shifted stimuli with features placed around the fibers’ best frequency. Individual AN fiber responses from anesthetized cats were measured for vowellike stimuli as a function of vowel and background-noise level. Similar to previous population studies, the STMP results show that spatiotemporal coding of vowel spectra (quantified using ALSR) is more robust than average-rate coding as vowel or background-noise level increases. The STMP results also suggest that the response phase transition near low-frequency formants may not be sharp enough to provide as robust encoding as previously suggested. This method of simulating spatiotemporal patterns from individual AN fiber responses may be particularly useful for studying the effects of sensorineural hearing loss because it is particularly difficult to obtain large populations of responses from animals with hearing loss. Preliminary data have been collected from cats with noise-induced hearing loss. [Work supported by NIH/NIDCD R03DC007348.]

Relative contributions of temporal and place pitch cues to fundamental frequency discrimination in cochlear implantees

The effect of the filter bank on fundamental frequency (F0) discrimination was examined in four Nucleus CI24 cochlear implant subjects for synthetic stylized vowel-like stimuli. The four tested filter banks differed in cutoff frequencies, amount of overlap between filters, and shape of the filters. To assess the effects of temporal pitch cues on F0 discrimination, temporal fluctuations were removed above 10 Hz in one condition and above 200 Hz in another. Results indicate that F0 discrimination based upon place pitch cues is possible, but just-noticeable differences exceed 1 octave or more depending on the filter bank used. Increasing the frequency resolution in the F0 range improves the F0 discrimination based upon place pitch cues. The results of F0 discrimination based upon place pitch agree with a model that compares the centroids of the electrical excitation pattern. The addition of temporal fluctuations up to 200 Hz significantly improves F0 discrimination. Just-noticeable differences using both place and temporal pitch cues range from 6% to 60%. Filter banks that do not resolve the higher harmonics provided the best temporal pitch cues, because temporal pitch cues are clearest when the fluctuation on all channels is at F0 and preferably in phase.

Detection thresholds for intensity increments in a single harmonic of synthetic Japanese macaque (Macaca fuscata) monkey coo calls.

There is evidence that Japanese macaques (Macaca fuscata) are extremely sensitive to dynamic changes in the relative amplitudes of coo call harmonics during discrimination tests. To verify this evidence using more controlled stimulus configurations, the authors examined threshold sensitivity of macaque monkeys to amplitude increments added to the standard level of coo call harmonics. Psychophysical threshold determination methods paralleled those used previously to determine macaque sensitivity to amplitude increments added to vowel-like stimuli. Variation was detectable although thresholds were elevated relative to those obtained with vowel-like stimuli in another investigation (C. G. Le Prell, A. J. Niemiec, & D. B. Moody, 2001). This elevation was probably a function of natural amplitude modulation in the standard stimuli.

Vowel discrimination in cats: thresholds for the detection of second formant changes in the vowel /epsilon/.

The ability of cats to discriminate changes in the second formant of the vowel /epsilon/ was examined across a range of stimulus levels. Cats were trained to press and hold down a lever to produce a pulsed train of the standard vowel /epsilon/, and to release the lever only when a variant of [epsilon] occurred. Six synthetic variants of /epsilon/ had the same first and third formants (F1 and F3), but with the second formant (F2) located between 1700 and 2000 Hz. All stimuli were tested at levels of 10, 30, 50, and 70 dB SPL. Average difference thresholds for changes in F2 (delta F2) of the vowel /epsilon/ ranged from 87 to 36 Hz across levels of 10 to 70 dB SPL, and were only slightly above those of humans. Further, the delta F2 values were lower than pure-tone delta F values in the same frequency range, whereas humans exhibit higher delta F2 values than pure-tone delta F values in the same frequency range. Changes in the second formant in a negative direction (downward F2 shifts) were also found to be more difficult to detect than upward F2 shifts. These results suggest that, compared to pure tones, cats are better able to discriminate small changes in more complex, vowel-like stimuli.

On the role of the fundamental frequency in vowel perception

Vowel identification tests were carried out using 200 synthesized vowel-like stimuli to examine the role of the fundamental frequency F0 in vowel perception. These stimuli were synthetic versions of the five Japanese vowels, /i/, /e/, /a/, /o/, and /u/, of which the F0 and/or the formant frequencies Fi (i = 1,2,3,4) were modified: ten F0 values were formed by adding n/3 Bark (n = 0,1,…,9) to the original F0. Four formant frequency sets were formed by adding m Bark (m = 0,1,2,3) to the original formant frequencies for each vowel. The results are the following: (1) perceived vowel height articulation shifts upward when the F0 shifts upward, while all formant frequencies remain the same: (2) this shift in vowel height is more distinct amid mid and low vowels than for high vowels; and (3) vowel height does not change when the F0 as well as all formant frequencies are shifted upward the same amount along the Bark scale. Further results, along with the hypothesis that a high F0 is regarded as the first formant in middle and low vowel perception, will be discussed.

Should ASR front-end be insensitive to fundamental frequency? (perceptual shift of formant position due to fine harmonic structure of voiced speech

Due to the fine harmonic structure of the source spectrum of voiced speech, most automatic speech analysis techniques estimate formant positions that systematically deviate from the resonant frequencies of the vocal tract. This property is in conflict with linear source-filter models of speech production and an effort is made to minimize the effect of the fundamental frequency F0 on the analysis result [e.g., Hermansky et al., Proceedings ICASSP-83 (IEEE, New York, 1983), pp. 778–780]. Our experiments show that the human auditory perception has properties similar to the above described properties of automatic analyses. The method of confusion scaling is applied in paired comparison of synthetic vowel-like stimuli with varying F0 and the second formant frequency F2. The results show that the perceptual estimate of formant position goes through a cycle of positive and negative deviations as the relative F2/F0 frequency goes from one integral number to the next. The position is estimated the most accurately when the formant peak falls either into a harmonic peak or if it lies right between two harmonic peaks. The maximal deviation of the estimate due to harmonics of the F0 from the 120-Hz neighborhood in the 2-kHz second formant region is about ± 0.5%. Our findings imply that when modeling the speech perception, as in the automatic speech recognition, a certain degree of sensitivity of the analysis technique to F0 might be beneficial.

Vowel perception: A neuroradiological localization of the perception of vowels in the human cortex

Whether the perception of vowels takes place in the right or left cerebral hemisphere, or is dependent on bilateral cortical processes, is of importance in our treatment of patients with language deficiences. To investigate this problem a phonetic test with a series of vowellike stimuli was administered to 68 right-handed adult patients with cerebral lesions, mainly of vascular origin. The results were compared to the results in a control group of 19 speech therapists. Second, a neuroradiologic method was developed in order to visualize the anatomic site of Wernicke's area in the left hemisphere on CT scans. This method formed the basis for an evaluation of the extent and localization of the patients' lesions in Wernicke's area. Of 46 patients with lesions in the left hemisphere, 19 had no perceptual disturbances and 27 had severe perceptual disturbances with lesions predominantly located in Wernicke's area. Twenty-two patients with lesions in the corresponding area in the right hemisphere showed no perceptual disturbances. The results of this investigation appear to indicate that the perception of vowels in right-handed persons is unilaterally located in Wernicke's area in the left hemisphere.

Speech coding in the auditory nerve: II. Processing schemes for vowel-like sounds.

Several processing schemes by which phonetically important information for vowels can be extracted from responses of auditory-nerve fibers are analyzed. The schemes are based on power spectra of period histograms obtained in response to a set of nine two-formant, steady-state, vowel-like stimuli presented at 60 and 75 dB SPL. One class of "local filtering" schemes, which was originally proposed by Young and Sachs [J. Acoust. Soc. Am. 66, 1381-1403 (1979)], consists of analyzing response patterns by filters centered at the characteristic frequencies (CF) of the fibers, so that a tonotopically arranged measure of synchronized response can be obtained. Various schemes in this class differ in the characteristics of the filter. For a wide range of filter bandwidths, formant frequencies correspond approximately to the CFs for which the response measure is maximal. If in addition, the bandwidths of the analyzing filters are made compatible with psychophysical measures of frequency selectivity, low-frequency harmonics of the stimulus fundamental are resolved in the output profile, so that fundamental frequency can also be estimated. In a second class of processing schemes, a dominant response component is defined for each fiber from a 1/6 octave spectral representation of the response pattern, and the formant frequencies are estimated from the most frequent values of the dominant component in the ensemble of auditory-nerve fibers. The local filtering schemes and the dominant component schemes can be related to "place" and "periodicity" models of auditory processing, respectively.

Sensorineural hearing loss and the discrimination of vowel-like stimuli.

It has been hypothesized that the wider-than-normal auditory bandwidths attributed to sensorineural hearing loss lead to a reduced ability to discriminate spectral characteristics in speech signals. In order to investigate this possibility, the minimum detectable depth of a spectral "notch" between the second (F2) and third (F3) formants of a synthetic vowel-like stimulus was determined for normal and hearing-impaired subjects. The minimum detectable notch for all subjects was surprisingly small; values obtained were much smaller than those found in actual vowels. An analysis of the stimuli based upon intensity discrimination within a single critical band predicted only small differences in performance on this task for rather large differences in the size of the auditory bandwidth. These results suggest that impairments of auditory frequency resolution in sensorineural hearing loss may not be critical in the perception of steady-state vowels.

Computerized Two-Dimensional Model for Finnish Vowel Identifications'

511 different vowel-like stimuli were produced using a computer and speech synthesizer OVE III B. Their first formant varied from 250 to 800 Hz and their second formant from 800 to 2 400 Hz, covering thus the formant frequencies of the eight Finnish vowels. The randomized stimuli were listened to by 32 young adults with normal hearing. The identifications were analyzed automatically with the computer and plotted on a two-dimensional plane as a function of F1 and F2. The resulting basic vowel identification chart demonstrates the areas where the stimuli were identified as a certain vowel. The chart offers the practical means for presentation and follow-up of individual articulatory and auditory capacities.

The ‘center of gravity’ effect in vowel spectra and critical distance between the formants: Psychoacoustical study of the perception of vowel-like stimuli

A two-formant synthetic vowel with closely spaced formants ( F 1 and F 2 being fixed) can be made perceptually similar to a single-formant stimulus with F ★ = (F 1 + F 2) 2 by adjusting the A 2 A 1 - amplitude ratio of the formants. The critical distance between the formants ( Δz c) that corresponds to the disappearance of this ‘center of gravity’ effect was found to be equal to 3.0–3.5 Bark. A close to continuous relation between A 2 A 1 and F ★ of a single-formant matching stimulus was found for stimuli with Δz < Δz c. A clear discontinuous relationship between A 2/ A 1 and F ★ was observed for stimuli with Δz > Δz c. For stimuli with Δz > Δz c, the formant amplitudes appeared to be of minor importance, the vowel quality being determined by the frequency locations of the formant peaks in the vowel spectrum. Formant peaks can be detected even if they are represented by very small spectral irregularities. Possible relations between peak extraction and ‘center of gravity’ effects are discussed.

Processing of place information in stop consonants

One of the basic questior, s that models of speech perception must answer concerns the conditions under which various cues will be extracted from a stimulus and the nature of the mechanisms which mediate this process. Two selective adaptation experiments were carried out to explore this question for the phonetic feature of place of articulation in both syllableinitial and syllable-final positions. In the first experiment, CV and VC stimuli were constructed with complete overlap in their second- and third-formant transitions. Despite this essentially complete overlap, no adaptation effects were found for a VC adaptor and a CV test series (or vice versa). In the second experiment, various vowel, vowel-like, and VC-like adaptors were used. The VC-like adaptors did have a significant effect on the CV category boundary, while the vowel and vowel-like stimuli did not. These results are interpreted within both one- and twolevel models of selective adaptation. These models are distinguished by whether selective adaptation is assumed to affect a single auditory level of processing or to affect both an auditory level and a later phonetic level. However, both models incorporate detectors at the auditory level which respond whenever particular formant transitions are present. These auditory detectors are not sensitive to the position of the consonant transition information within the syllable.

Elicitation of the acoustic reflex with synthetically produced consonant and vowel-like stimuli

Acoustic reflexes were elicited from seven normal-hearing subjects with synthesized isolated consonant and vowel-like stimuli. Stimulus intensity and duration were controlled so that equivalent amplitudes and specific durations were produced. The recorded stimulus items were presented over an earphone to the subject's right ear, while susceptance changes in response to bilateral acoustic reflex contractions were monitored in the subject's contralateral (left) ear. Analysis of variance was performed on the data for acoustic reflex thresholds. No significant difference between isolated consonant and vowel-like reflex threshold values was demonstrated. The main effects of stimulus duration and subjects, and the two-way interaction of phoneme type by subject were found to be significant. [Work supported by NIH.]