Perceptual categorization and consistency of synthesized (r-w) continua by adults, normal children and (r)-misarticulating children

The purpose of this study was to determine if children who misarticulate (r) differ from normal children and adults in the perception of sound features that are produced correctly and incorrectly. Children with normal articulation, children who produced (r) misarticulations, and adults listened to synthesized child and adult (r-w) continua in two separate sessions, and to an adult (b-w) control continuum in one session. Perception was evaluated on the basis of measures of phonetic boundary location and the consistency of response to each stimulus in a continuum. The (r)-misarticulating children were found to be significantly less consistent than child and adult controls in responding to the (r-w) stimuli. Moreover, consistency scores were significantly higher for the adult continuum than for the child continuum. The performance of children was different from that of adults. Due to inconsistent performance, boundaries could not be computed for (r)-misarticulating children, but it was found that the boundaries for children in the control group were closer to the (r)-end of the continuum than those for adults. In the case of the (b-w) continuum, it was found that (r)-misarticulating children were significantly less consistent than adults. The phonetic boundaries of children were significantly closer to the (b)-end of the continuum than the boundary for adults. Thus, the results reveal that variability in stimulus response was influenced primarily by the productive ability of the subjects, whereas differences in stimulus categorization were influenced by the age of the subjects. The perceptual variability was most clearly reflected by responses to stimuli produced incorrectly, whereas categorization differences extended to sounds produced correctly.     

Effect of formant transition rate on the differentiation of synthesized child and adult (w) and (r) sounds

The purpose of this research was to assess the perceptual effects of a range of second (F2) and third (F3) formant transition rates that occur naturally in the production of /w/ and /r/ by children and adults. Synthesized CV continua that varied in the second (F2) and third (F3) formant onset frequencies between values appropriate for /w/ and /r/ were used as stimuli. Subjects participated in four experimental conditions that involved changing the rate of transition of either F2 or F3 by varying the duration of the transition between the glide onset and an /eI/ vowel nucleus for child and adult stimuli. In each condition, the transition rates of the /w/-endpoint stimulus, the /r/-endpoint stimulus, and at least one of the midpoint stimuli were varied across values appropriate for /w/ and /r/. Mean ratings of the stimuli were compared to test the predictions that /w/ perception increases with increased F2 transition rate and increases with decreased F3 transition rate. The results were as follows: (a) one of the 10 comparisons for the child F2 stimuli was significant, but it involved a change opposite to the predicted direction; (b) four of the six comparisons for the child F3 stimuli were significant, but they involved changes opposite to the predicted direction; (c) 14 comparisons of adult F2 stimuli were significant, but 12 of these 14 comparisons involved changes opposite to the predicted direction; and (d) four of the six comparisons for the adult F3 stimuli were significant, but none of them involved changes in the predicted direction. Only one of all the comparisons involved a significant change in rating between the /w/ and /r/ categories.     

Perception of distorted "R" sounds in the synthesized speech of children and adults

Adult and Child manifolds were generated by synthesizing 5 X 5 matrices of /Cej/ type utterances in which F2 and F3 frequencies were systematically varied. Manifold stimuli were presented to 11 graduate-level speech-language pathology students in two conditions: (a) a rating condition in which stimuli were rated on a 4-point scale between good /r/ and good /w/; and (b) a labeling condition in which stimuli were labeled as "R," "W," "distorted R," or "N" (for none of the previous choices). It was found that (a) stimuli with low F2 and high F3 frequencies were rated 1.0-1.4; those with high F2 and low F3 frequencies were rated 3.6-4.0 were labeled as "R"; (c) none of the Child manifold stimuli were labeled as distorted "R" and one of the Adult manifold stimuli approached a level of identification that approached the percentage of identification for "R" and "W"; and (d) rating and labeling tasks were performed with a high degree of reliability.     

Identification of synthetic, voiced stop-consonants by hearing-impaired listeners

The identification of synthetic, voiced stop consonants changing in place of articulation was investigated in 26 normal-hearing listeners and 16 hearing-impaired listeners who reported difficulties in speech understanding, but attained high speech discrimination scores on standard test materials. Two place continua (each containing 13 stimuli changing from /ba/ to /da/ and /ga/) were employed: a continuum in which place was cued by an initial noise burst and formant transitions, and a continuum in which place was cued by formant transitions alone. All normal listeners exhibited categorical and highly consistent identification for both stimulus continua. For the hearing-impaired listeners, identification was somewhat less consistent for the burst-and-transition stimuli, and much poorer for the transition-only stimuli. Similar differences were observed when comparing the responses of 4 subjects with a unilateral hearing loss when the stimuli were presented to their normal and impaired ears, and 2 subjects with a unilateral, sudden-onset hearing loss with recovered pure-tone sensitivity.     

Comparison of speech-language pathologists' and naive subjects' identification of synthesized /r-w/ continua

Two 9-step continua varying in F2 and F3 frequencies between exemplary /r/ and /w/ were synthesized to represent child and adult talkers. Stimuli from the full and truncated versions of the continua were presented to naive subjects in Experiment 1 and to speech-language pathologists in Experiment 2. Shifts from full range continua category boundaries occurred in Experiment 1 for both truncated "R" and truncated "W" conditions and in a direction opposite to the truncated end of the continuum. The results of Experiment 2 indicated that the full-range category boundaries for speech-language pathologists differed from those for naive subjects but that the boundary shift for truncated "R" was as great as that for naive subjects and the boundary shift for truncated "W" was greater than that for naive subjects. These findings indicate that speech-language pathologists are more likely than naive individuals to judge ambiguous /r/ sounds as "W" and that the phonetic judgments of speech-language pathologists about /r/ sounds are no more stable than those of naive subjects.     

An exploratory study into perception of acoustic speech cues by hearing-impaired adults

The aims of the present study were to assess discrimination and identification based on two classes of acoustic cue by adults with acquired sensorineural hearing impairment. Eight hearing-impaired and eight normally hearing adults were asked to identify and discriminate two different sets of speech stimuli. A plosive voicing continuum (coat/goat) varied in voice onset time. The plosive place of articulation continuum (date/gate) varied in burst spectra and second formant transition. Subjects were tested in the unaided condition with the exception of one hearing-impaired subject for whom speech was completely inaudible without a hearing aid. There was no significant between-group difference in discrimination or identification of the voicing contrast. There was no significant between-group difference in identification of stimuli varying by place of articulation. However, three of the eight hearing-impaired subjects were very poor at identification. The hearing-impaired subjects also showed significantly impaired place of articulation discrimination. Both measures were significantly correlated with threshold at 2000 Hz. The results support the view that hearing impairment can have different effects on perception of different acoustic contrasts and on different psychophysical tasks.     

Frequency discrimination ability and stop-consonant identification in normally hearing and hearing-impaired subjects

Identification of place of articulation in the synthesized syllables /bi/, /di/, and /gi/ was examined in three groups of listeners: (a) normal hearers, (b) subjects with high-frequency sensorineural hearing loss, and (c) normally hearing subjects listening in noise. Stimuli with an appropriate second formant (F2) transition (moving-F2 stimuli) were compared with stimuli in which F2 was constant (straight-F2 stimuli) to examine the importance of the F2 transition in stop-consonant perception. For straight-F2 stimuli, burst spectrum and F2 frequency were appropriate for the syllable involved. Syllable duration also was a variable, with formant durations of 10, 19, 28, and 44 ms employed. All subjects' identification performance improved as stimulus duration increased. The groups were equivalent in terms of their identification of /di/ and /gi/ syllables, whereas the hearing-impaired and noise-masked normal listeners showed impaired performance for /bi/, particularly for the straight-F2 version. No difference in performance among groups was seen for /di/ and /gi/ stimuli for moving-F2 and straight-F2 versions. Second-formant frequency discrimination measures suggested that subjects' discrimination abilities were not acute enough to take advantage of the formant transition in the /di/ and /gi/ stimuli.     

Mismatch negativity to acoustic differences not differentiated behaviorally

Mismatch negativity (MMN) reportedly reflects the neurophysiologic detection of acoustic differences, rather than the phonemic categorization of speech sounds. The purpose of the present study was to determine if it is elicited by speech contrasts that are acoustically different but are not differentiated by listeners in behavioral tasks. Experimental stimuli were drawn from a synthetically generated continuum that varied in place of articulation from /da/ to /ga/. Contrasts used to elicit MMN were (a) the continuum endpoints, (b) the two-step contrast that straddled each listener's categorical boundary, and (c) a within-category contrast that was not behaviorally differentiated by any of the listeners. MMN responses were elicited by all three experimental contrasts. It appears that MMN may be an index of the neurophysiology underlying the ability or inability to discriminate the acoustic parameters necessary for speech perception, rather than a neurophysiologic correlate of behavioral speech discrimination ability. However, limitations involving identification of MMN in the responses of individual listeners confounded this interpretation.     

Discrimination of intonation contours by adolescents with cochlear implants

Abstract

Objective: Differences in fundamental frequency (F0) contour peak alignment contribute to the perception of pitch accents in speech intonation. The present study assessed the discrimination of differences in F0 contour peak alignment by adolescent users of cochlear implants (CIs). Design: In Experiment 1, subjects discriminated between rise-fall F0 contours located early in the syllable and those aligned late. Recorded utterances with manipulated F0 were used as stimuli and all subjects wore a unilateral CI. In Experiment 2, bilaterally-implanted subjects repeated Experiment 1 in the bilateral condition. Study sample: Twenty-one CI users aged 12–21 years participated. A normally-hearing control group (n = 20) also completed Experiment 1. Results: Listeners with normal hearing (NH) could discriminate between F0 peaks differing by 80 ms or more. Results varied among the CI users, with only four users displaying a pattern of results similar to that of the NH listeners. Sixteen CI users responded inconsistently or at chance levels (p > 0.05; binomial test). Ten CI users who were bilaterally implanted completed the tests in unilateral and bilateral listening conditions. Conclusions: Results suggest that CI users may have difficulty discriminating between F0 alignment and that use of bilateral implants did not provide an advantage to discrimination.     

Effect of formant frequency onset variation on the differentiation of synthesized /w/ and /r/ sounds

The purpose of this study was to assess the use of psychophysical transformations for analyzing the differentiation of /w/ and /r/ sounds of children and adults. Stimuli from Adult and Child manifolds, consisting of 25 synthesized /Cej/-type utterances with different F2 and F3 onset frequencies, were presented in random order to eight naive subjects. Subjects rated the stimuli on a four-point scale between good /r/ and good /w/. Correlations between mel transformations and Bark transformations of the F3-F2 differences among the stimuli and their percent /r/ responses were close to or greater than .90. Predictions of percent /r/ responses derived from regression analyses based on mel transformations and Bark transformations of F3-F2 differences among stimuli indicated that some sounds identified as /w/ for /r/ substitutions could be differentiated from /w/ sounds. The category boundaries between /r/ and /w/ were estimated to be 5.0 Bark for adult stimuli and 5.7 Bark for child stimuli.     

Perception of synthetic visual consonant-vowel articulations

Synthetic speech-like articulations were presented to adult subjects via the visual modality, following the classic categorical perception experimental paradigm (Liberman, Harris, Hoffman, & Griffith, 1957). Animations were generated on a computer-based graphics system. Stimuli consisted of representations of the syllables /ba/, /va/, and /wa/; as well as 6 linearly interpolated intermediate stimuli between each of the possible exemplar pairs, resulting in three 8-item continua. Three sets of observations were obtained for these stimuli. First, for each continuum, labeling data were obtained in which the subject assigned one or the other exemplar label to each of the stimuli. Next, ABX discrimination data were obtained for each continuum. In the final task, subjects assigned a rating of one through nine to each animation indicating the extent to which it was like the exemplar syllables. Although the labeling functions showed rather abrupt transitions from one response category to the other, the peaks in the discrimination functions did not coincide with the category boundaries. Further, the mean rating functions were relatively linear, and the distribution of rating responses revealed unimodal distributions whose peak locations differed depending on the stimulus.     

The role of spectral and temporal cues in vowel identification by listeners with impaired hearing.

This study examined the use of duration and formant frequency in the labeling of synthetic CVC stimuli forming a beet-bit continuum. Durational and F2 frequency cues to vowel identity varied systematically across stimuli. Subjects with normal hearing tended to rely primarily on F2 frequency in vowel labeling, whereas subjects with impaired hearing relied less on F2 information. This group difference was observed even for stimuli with large F2 differences, which were easily discriminated by all subjects. The effect of vowel duration on labeling was similar for both groups, with long-duration stimuli receiving more "beet" responses than short-duration stimuli across the F2 range. Psychoacoustic measures of frequency resolution and temporal resolution were poor predictors of a subject's use of formant information and duration information in labeling.     

Perception of voiceless fricatives by children with a functional articulation disorder

The two studies presented here examine the relationship between speech perception and speech production errors in children who have a functional articulation disorder. In both experiments, speech perception was assessed with a word identification test, based upon a synthesized continuum of speech stimuli, contrasting the specific phonemes that were associated with production errors in our sample of articulation-disordered subjects. Experiment 1 required subjects to identify words that contrasted the phonemes /s/ and /S/. In this test, adults, normal speaking 5-year-olds, and some articulation-disordered 5-year-olds identified the words seat and sheet appropriately and reliably. However, a subgroup of articulation-disordered children were unable to identify the test stimuli appropriately. Experiment 2 required a second group of subjects to identify words that contrasted the phonemes /s/ and /theta/. Although both adults and normal speaking children responded appropriately to the words sick and thick, in this test, none of the articulation-disordered children was able to identify these words appropriately. It is concluded that, for a subgroup of children who have a functional articulation disorder, production errors may reflect speech perception errors.     

Children with specific language impairments perceive speech most categorically when tokens are natural and meaningful.

PURPOSE: To examine perceptual deficits as a potential underlying cause of specific language impairments (SLI). METHOD: Twenty-one children with SLI (8;7-11;11 [years;months]) and 21 age-matched controls participated in categorical perception tasks using four series of syllables for which perceived syllable-initial voicing varied. Series were either words or abstract nonword syllables and either synthesized or high-quality edited natural utterances. Children identified and discriminated (a) digitally edited tokens of naturally spoken "bowl"-"pole", (b) synthesized renditions of "bowl"-"pole", (c) natural "ba"-"pa", and (d) synthetic "ba"-"pa". RESULTS: Identification crossover locations were the same for both groups of children, but there was modestly less accuracy on unambiguous endpoints for children with SLI. Planned comparisons revealed these effects to be limited to synthesized speech. Children with SLI showed overall reduced discrimination, but these effects were limited to abstract nonword syllables. CONCLUSION: Overall, children with SLI perceived naturally spoken real words comparably to age-matched peers but showed impaired identification and discrimination of synthetic speech and of abstract syllables. Poor performance on speech perception tasks may result from task demands and stimulus properties, not perceptual deficits.     

The frequency-modulation following response in young and aged human subjects

The frequency-modulation following response (FMFR) is a steady-state evoked response which may be a neural correlate of frequency discrimination. Aged subjects with normal hearing have abnormal frequency discrimination for low carrier frequencies and thus it might be predicted that aged individuals would have reduced FMFR amplitudes compared to young subjects. In this study, FMFR amplitudes were measured for frequency-modulated sinusoids with a carrier frequency of 0.5 kHz (80 dB SPL). In Experiment 1, the modulation depth was held constant (80%) and the modulation rate was varied (4-38 Hz), whereas in Experiment 2 the modulation rate was held constant (38 Hz) and the modulation depth was varied (0-80%). Aged subjects had significantly larger FMFR amplitudes than young subjects for certain stimulus parameters, although individual variability was large. Such results would not be predicted given previous data regarding frequency discrimination, but are consistent with several reports of larger-than-normal amplitudes of middle latency and late responses in aged subjects.     

Speech perception in adult subjects with familial dyslexia.

Speech perception was investigated in a carefully selected group of adult subjects with familial dyslexia. Perception of three synthetic speech continua was studied: /a/-/e/, in which steady-state spectral cues distinguished the vowel stimuli; /ba/-/da/, in which rapidly changing spectral cues were varied; and /sta/-/sa/, in which a temporal cue, silence duration, was systematically varied. These three continua, which differed with respect to the nature of the acoustic cues discriminating between pairs, were used to assess subjects' abilities to use steady state, dynamic, and temporal cues. Dyslexic and normal readers participated in one identification and two discrimination tasks for each continuum. Results suggest that dyslexic readers required greater silence duration than normal readers to shift their perception from /sa/ to /sta/. In addition, although the dyslexic subjects were able to label and discriminate the synthetic speech continua, they did not necessarily use the acoustic cues in the same manner as normal readers, and their overall performance was generally less accurate.     

Perception/production relationships in the development of the vowel duration cue to final consonant voicing

The purposes of this study were to assess: (a) the development of identification in children for the vowel duration cue to final consonant voicing and (b) the perception/production relationships in children for the vowel duration cue to final consonant voicing. Subjects were 30 children divided equally into three age groups, and 10 adults. Productions consisted of 15 repetitions of two target syllables (beet, bead) analyzed acoustically for vowel duration. From these were calculated category boundary, category separation, and variability in production for each subject. Perceptual data were collected using a synthesized speech continuum that varied vowel duration. Identification responses were used to calculate category boundary, category separation (slope/boundary width) and variability (response consistency) for each subject. Mean percentage correct discrimination was derived by using two-step and three-step two-pair same-different paradigms. The results were as follows: (a) category boundary and category separation in production were adult-like by 8 years of age, (b) variability in production was not adult-like by 10 years of age, (c) perception categorization (category boundary and category separation) was adult-like at 5 years of age, (d) perceptual consistency was not adult-like until 10 years of age, (e) percentage correct discrimination was not adult-like by 10 years of age, (f) correlations between comparable perception and production measures were nonsignificant, and (g) a pairwise comparisons analysis indicated that perception was consistently more advanced than production.     

Production and perception of distortion in word-initial friction duration

The purpose of the present study was to investigate (a) the distortion in production of word-initial friction duration in fricative /?/, and (b) the perceptual discrimination between typical (normal) and atypical (prolonged or lengthened) friction duration. In the first experiment 80 school aged children pronounced word /?uma/, 40 of them have had typical friction duration, 15 shortened and 25 lengthened duration of /?/, and all of them did not have any distortion in segment /uma/. Measurement and analysis of duration of both segments /?/ and /uma/ were performed after segmentation of speech stimuli /?uma/. In second experiment friction duration of /?/ has been synthesized in 17-step continuum from 135 ms to 311 ms in equal steps of 11 ms, friction duration of 223 ms has accepted as average typical value, and segment /uma/ was unchanged. Six speech-language pathologists (SLPs) listened to the synthesized stimuli /?uma/ and performed perception task – distortion contrast in two sessions. They discriminated the shortened and lengthened duration in comparison to typical duration of fricative /?/ and results were interpreted by identification functions. The results of the first experiment indicate that distortion in production of friction duration of /?/ does not affects duration of the other segment /uma/ in word /?uma/, and that these two durations are non-correlated. In perception of distortion in friction duration SLPs show high inter-listeners agreement and their identification functions showed very good discrimination in typical/atypical friction duration.Learning outcomes: Readers will be able to: (1) understand the nature of distortion in friction duration in terms of (a) the problems in control of articulators and (b) the fact that it is an inherent feature independent to phonetic environment; (2) become familiar with the identification functions in discrimination of typical/atypical articulation of an acoustic cue; and (3) understand the difference between phonetic contrast and distortion contrast in speech perception.     

Voice Discrimination by Adults with Cochlear Implants: the Benefits of Early Implantation for Vocal-Tract Length Perception

Cochlear implant (CI) users find it extremely difficult to discriminate between talkers, which may partially explain why they struggle to understand speech in a multi-talker environment. Recent studies, based on findings with postlingually deafened CI users, suggest that these difficulties may stem from their limited use of vocal-tract length (VTL) cues due to the degraded spectral resolution transmitted by the CI device. The aim of the present study was to assess the ability of adult CI users who had no prior acoustic experience, i.e., prelingually deafened adults, to discriminate between resynthesized “talkers” based on either fundamental frequency (F0) cues, VTL cues, or both. Performance was compared to individuals with normal hearing (NH), listening either to degraded stimuli, using a noise-excited channel vocoder, or non-degraded stimuli. Results show that (a) age of implantation was associated with VTL but not F0 cues in discriminating between talkers, with improved discrimination for those subjects who were implanted at earlier age; (b) there was a positive relationship for the CI users between VTL discrimination and speech recognition score in quiet and in noise, but not with frequency discrimination or cognitive abilities; (c) early-implanted CI users showed similar voice discrimination ability as the NH adults who listened to vocoded stimuli. These data support the notion that voice discrimination is limited by the speech processing of the CI device. However, they also suggest that early implantation may facilitate sensory-driven tonotopicity and/or improve higher-order auditory functions, enabling better perception of VTL spectral cues for voice discrimination.