• Volume/Page
  • Keyword
  • DOI
  • Citation
  • Advanced
   
 
 
 

Journal of the Acoustical Society of America

SECTION LISTING

BROWSE VOLUMES

Year Range: 
Search Issue | RSS Feeds RSS
Previous Issue Next Issue

Apr 1982

Volume 71, Issue S1, pp. S1-S113

Return to All Sections
back to top
RSS Feeds
back to top Session AA. Speech Communication VI: Speech Perception and Production in the Hearing Impaired (Poster Session)
Contributed Papers
FREE

CV detection and identification thresholds for speech/language/learning‐disordered listeners (A)

Lu Ann B. Clifton and Lois L. Elliott

J. Acoust. Soc. Am. Volume 71, Issue S1, pp. S57-S57 (1982); (1 page)

Online Publication Date: 12 Aug 2005

Full Text: | Download PDF

Show Abstract
This report is a continuation of one by the same authors [J. Acoust. Soc. Am. Suppl. 1 70, S32 (1981)] at the 102nd ASA meeting and will emphasize detection and identification of synthesized CV stimuli [bɑ, dɑ, gɑ]. Responses of children with articulation disorders and of learning‐disabled adults deviated from normal performance [L. L. Elliott et al., Percep. Psychophys. 30 (5), 411–416 (1981)]. Poorer detection thresholds, poorer identification thresholds, and aberrant slopes of the identification functions occurred; however, there were large individual differences. Use of standard audiological tests could not have predicted the experimental performance of these listeners. [Supported, in part, by NSF.]
FREE

Improvement in consonant recognition with amplification: Correlations with PTA and PTCs (A)

M. F. Dorman and M. Hannley

J. Acoust. Soc. Am. Volume 71, Issue S1, pp. S57-S57 (1982); (1 page)

Online Publication Date: 12 Aug 2005

Full Text: | Download PDF

Show Abstract
Presbyacusic listeners with moderate sloping sensorineural bearing impairment were presented the California Consonant Test in three conditions: unaided; with flat frequency response aid; and with high‐frequency emphasis hearing aid. Frequency selectivity was assessed using the metric Q1 octave, for a simultaneous pure tone masking paradigm. Three distinct error patterns were found when the data from the unaided and aided conditions were compared: (1) errors on a phone disappeared with amplification (e.g./tʃ); (2) errors changed from one phone to another with amplification (e.g./for/s/); and (3) the errors were the same and were not eliminated by amplification (for the stop consonants). In the latter case, the best correlation of PTA for any set of frequencies and improvement in stop consonant recognition was −0.35; the best correlation of Q1 octave and improvement in stop recognition under aided conditions was 0.63. Even though Q1 octave was a better predictor of improvement than was the PTA, the range of improvement with similar Q scores was considerable.
FREE

Accounting for errors in /s/ identification by the hearing impaired (A)

M. F. Dorman, M. Hannley, and L. J. Raphael

J. Acoust. Soc. Am. Volume 71, Issue S1, pp. S57-S57 (1982); (1 page)

Online Publication Date: 12 Aug 2005

Full Text: | Download PDF

Show Abstract
Previous research with hearing impaired listeners has reported a number of substitution errors for /s/, e.g., /ʃ/, /θ/, /f/, and /t/. The purpose of the research reported here was to determine the conditions under which each of the errors occurs and thus to determine the predictability of errors. Hearing‐impaired listeners were first presented /s/‐noise in isolation to determine threshold and then were presented a test sequence at several levels above /s/ threshold. The confusion errors for /s/ were found to group into reasonable patterns as a function of /s/ threshold and presentation level.
FREE

Development of a portable speech processor for an auditory implant subject: Choice of algorithm and role of learning (A)

Les E. Atlas, Matthew K. Herndon, Robert L. White, and Laurel J. Dent

J. Acoust. Soc. Am. Volume 71, Issue S1, pp. S57-S57 (1982); (1 page)

Online Publication Date: 12 Aug 2005

Full Text: | Download PDF

Show Abstract
Several real‐time portable speech processing algorithms for an ditory implant subject [Simmons et al., Acta Otolaryngol. 87, 170–175 (1979)] have been evaluated. The algorithms, designed in consideration of data collected on loudness and pitch for this implant subject, included various combinations of (1) automatic gain control, (2) instantaneous compression, (3) emphasis of high‐frequency energy if unvoiced sounds are detected, (4) differentiation following the instantaneous compression, and (5) encoding zero crossings as pulses with amplitude derived from the speech envelope. Taped speech tests were processed by each of the algorithms and presented to the subject via a single modiolus‐implanted electrode. Findings include the observations that conversion of zero crossings to pulses led to improved perception of the question/statement distinction and that initial consonant identifications were facilitated by the differentiation following the compression. Having thus determined the best combination of the above‐mentioned processing steps, we incorporated this combination into a portable speech processor. The effect of daily use of the portable processor on speech discrimination and identification will be reported. [Work supported by NIH.]
FREE

Comparison of four speech‐processing algorithms in one cochlear implant subject (A)

Matthew K. Herndon, Les E. Atlas, F. Blair Simmons, and Robert L. White

J. Acoust. Soc. Am. Volume 71, Issue S1, pp. S57-S57 (1982); (1 page)

Online Publication Date: 12 Aug 2005

Full Text: | Download PDF

Show Abstract
Four different methods of encoding speech for electrical stimulation were compared in a single cochlear implant subject using taped speech material. Thus the effectiveness of the different speech processing strategies could be compared without subject‐to‐subject differences. A baseline algorithm presented only information about the amplitude envelope of the acoustic speech signal. A second algorithm added fundamental frequency information; improved comprehension of final consonants in monosyllabic words was seen, apparently due to increased ability to distinguish voiced and unvoiced consonants. A third algorithm consisted of simple log compression of the analog signals; comprehension was somewhat better than with the baseline algorithm, but it is unclear which features of speech were responsible for this performance. The fourth algorithm extracted formant frequencies and mapped them onto the lower usable frequency range [Simmons et al., Acta Otolaryngol. 87, 170–175 (1979)]. This algorithm showed the best overall comprehension, especially for initial and final consonants of monosyllabic words. [Work supported by NIH.]
FREE

An investigation of the English vowels of intelligible but non‐normal profoundly deaf adult male speakers (A)

Linda Rowe

J. Acoust. Soc. Am. Volume 71, Issue S1, pp. S57-S58 (1982); (2 pages)

Online Publication Date: 12 Aug 2005

Full Text: | Download PDF

Show Abstract
The English vowels of ten intelligible but non‐normal profoundly deaf adult male speakers were analyzed using computer techniques to determine ferment frequency values. Spectrographic analysis of the same data was compared to the computer analysis to assure that the computer techniques were applicable to the non‐normal speech of deaf speakers. The acoustic data from the deaf speakers was compared to the same data obtained from normal hearing and speaking adult males. Upon completion of the initial research, it is likely that the deaf speakers will miss the acoustic vowel targets. It is reasonable to assume that the deaf speakers are not able to place their articulators accurately enough to achieve the correct targets, and that they are unable to correct the inaccurate articulator placement through auditory monitoring. It will be of interest to examine the relationship between the acoustic data and the physiological correlates and the impact this information should have on the remediation strategies used to improve the speech of profoundly deaf speakers.
FREE

Teaching speech to the deaf using computer‐assisted instruction (A)

M. J. Osberger, R. P. Lippmann, R. D. Kent, and C. S. Watson

J. Acoust. Soc. Am. Volume 71, Issue S1, pp. S58-S58 (1982); (1 page)

Online Publication Date: 12 Aug 2005

Full Text: | Download PDF

Show Abstract
At the 98th meeting of the Acoustical Society, we presented a preliminary report on a recently developed computer‐based speech training aid for the deaf [Lippmann and Watson, J. Acoust. Soc. Am. Suppl. 1 66, S13 (1979)]. The aid evaluates a child's utterances using a template comparison similar to that used in speaker‐dependent word recognition schemes. Characteristics determined from a real‐time acoustic analysis of a child's utterance are compared to those of a previously recorded intelligible utterance. The aid is being evaluated in a systematic speech training program with a group of deaf children, 8 to 14 years of age. This report will present preliminary data on the initial clinical trials with the aid. The data indicate that the aid is at least as effective as a teacher in performing the speech drills under study. The children have learned to interpret the feedback, enjoy using the computer, and even after six months of intensive training, remain highly motivated to perform the computerized speech drills. The degree of acoustic variation in the children's productions during the training sessions and its effect on the accuracy of the computer evaluations will also be discussed. [Research supported by NIH/NINCDS‐NS16247.]
FREE

Identification of an emphasized consonant of a monosyllable in hearing‐impaired subjects and its application to a hearing aid (A)

Hiroshi Ono, Tsutomu Okasaki, Shoichiro Nakai, and Hidenobu Harasaki

J. Acoust. Soc. Am. Volume 71, Issue S1, pp. S58-S58 (1982); (1 page)

Online Publication Date: 12 Aug 2005

Full Text: | Download PDF

Show Abstract
It is generally known that sensorineural hearing impairment has not only decreased hearing ability but also disorders in various hearing function and for this reason, with the conventional hearing aid which merely loudens sound, failure to catch or miss hearing a consonant often occurs. We have thought to improve audibility by separating a consonant and vowel of a monosyllable then synthesizing each to form an emphasized sound. This synthesized monosyllable was tested for speech discrimination score by 20 sensorineural hearing‐impaired subjects. The results show that in the case of voiceless plosives such as /pa/, /ta/, and /ka/, when amplitude of a consonant was increased by 10–20 dB and a silent pause of 30–60 ms was inserted between a consonant and vowel to prevent masking by a following vowel, recognition improved by 46.7%. The improvement is remarkable in the case of /ka/ where original sound was recognized by 40% and synthesized sound by all the subjects. On the other hand, in the case of nasal sound such as /ma/ and /na/, by repetition of consonant and of amplified consonant the audibility improved to 70% from the original recognition of 22.5%. Now we are developing a computer hearing aid which gives the best synthesized monosyllable for bearing‐impaired subjects based on speech analysis.
FREE

Impaired‐ and normal‐hearing listeners' voicing perception of final fricatives in cue‐adjusted syllables (A)

S. Revoile, J. M. Pickett, and L. Holden

J. Acoust. Soc. Am. Volume 71, Issue S1, pp. S58-S58 (1982); (1 page)

Online Publication Date: 12 Aug 2005

Full Text: | Download PDF

Show Abstract
In the spoken syllables /d∧z/ versus /d∧s/ and /d∧v/ versus /d∧f/, vowel duration differences were reduced, frications deleted, and vowel offsets exchanged or deleted. These adjustments were performed digitally on ten different tokens of each syllable. The voicing cues were adjusted in various combinations to yield different test conditions for identification of the syllables. Hearing‐impaired (N = 24) and normal‐hearing (N = 11) listeners were tested several times under each condition. For both listener groups, fricative voicing perception was generally unaffected by deletion of the frications from the syllables. The vowel duration adjustments (lengthening vowels that preceded voiceless frication, shortening vowels preceding voiced) significantly reduced fricative voicing perception for about half of the hearing‐impaired listeners; none of the normal‐hearing listeners was significantly affected by the vowel duration adjustment. When the vowel offsets (final seven pitch periods) were exchanged between voiced and voiceless syllables with vowel durations adjusted and frication deleted, voicing perception went with the exchanged vowel offsets. Deletion of the vowel offsets substantially reduced voicing perception.
FREE

Effects of modifying the time‐varying amplitude patterns on perception of speech by hearing‐impaired and normal listeners (A)

Larry K. Henrickson

J. Acoust. Soc. Am. Volume 71, Issue S1, pp. S58-S58 (1982); (1 page)

Online Publication Date: 12 Aug 2005

Full Text: | Download PDF

Show Abstract
The acoustical patterns of speech may be analyzed into three primary components: a fine‐temporal pattern, a gross‐temporal pattern, and a time‐varying amplitude pattern. Previous research has supported the importance of both fine‐ and gross‐temporal cues for speech perception. Braida et al. [ASHA Monographs 19 (1979)] reviewed amplitude compression, and indicated that the importance of time‐varying amplitude patterns was not well understood. The present research utilized altered versions of the California Consonant Test to examine perception by hearing impaired and normals. The time‐varying amplitude patterns had been systematically altered to simulate the effects of several types of single‐ and multichannel compressors. The effects of frequency response, degree of compression, and number of channels have been statistically analyzed using ANOVA. Results for the hearing impaired showed significant improvements in consonant differentiation for severe rather than mild compression ratios, and for compression preceded by high‐frequency emphasis. Multichannel designs did not perform better than optimal single‐channel designs. The time‐varying amplitude pattern seems of little importance for the speech intelligibility.
FREE

Influence of context on measurement of hearing‐impaired speakers' intelligibility (A)

Ronald Sitler, Nicholas Schiavetti, and Dale E. Metz

J. Acoust. Soc. Am. Volume 71, Issue S1, pp. S58-S59 (1982); (2 pages)

Online Publication Date: 12 Aug 2005

Full Text: | Download PDF

Show Abstract
Previous research has indicated that hearing‐impaired talkers' intelligibility scores are better when sentences are used than when word lists are used as speech material in word identification tests. The speech intelligibility of 20 hearing‐impaired talkers was measured with word identification tests using isolated words (W‐22 monosyllables) and words in sentence context (CID sentences). Analysis of individual talkers' intelligibility data revealed that sentence intelligibility scores were higher than word intelligibility scores only for the better talkers and that no differences were apparent between sentence and single word intelligibility for the poorer talkers. These findings agree with the results of research with normal speech that has been degraded in intelligibility by noise or filtering and indicate that an interaction may exist between context and overall intelligibility in which only speech that has a certain degree of overall intelligibility may show further intelligibility improvement with increased contextual clues. [Work supported by U.S. Department of Education.]
FREE

Multiband syllabic compression for severely impaired listeners (A)

S. De Gennaro, L. D. Braida, and N. I. Durlach

J. Acoust. Soc. Am. Volume 71, Issue S1, pp. S59-S59 (1982); (1 page)

Online Publication Date: 12 Aug 2005

Full Text: | Download PDF

Show Abstract
Speech reception experiments were conducted to evaluate the benefit of multiband syllabic compression for listeners with severe sensorineural hearing impairments. Three listeners with congenital hearing loss characterized by flat audiograms and dynamic ranges of 20–30 dB were tested on three compression systems and one linear amplification system. The compression systems placed progressively larger amounts of speech energy within the listener's residual dynamic range by raising to audibility and compressing 25%, 50%, and 90% of the short‐term input amplitude distribution in each of 16 frequency bands. The comparison linear system was defined by adjusting six octave‐wide bands of speech to comfortable levels. System performance was evaluated with nonsense CVC syllables presented at a constant input level and spoken by two talkers. Extensive training was provided to ensure stable performance. The results were notably speaker dependent, with compression consistently providing better performance for one, linear amplification for the other. Averaged over speakers, however, there was no net advantage for any of the compression systems for any listener. The use of high compression ratios and large input ranges tended to degrade perception of initial consonants and vowels. Under some conditions, however, final consonant scores were higher with compression than with linear amplification. Detailed summaries of system performance and analyses of consonant and vowel confusions will be presented. [Work supported by NIH.]
FREE

Sign language perception and linguistic experience (A)

Rachel Mayberry and Susan D. Fischer

J. Acoust. Soc. Am. Volume 71, Issue S1, pp. S59-S59 (1982); (1 page)

Online Publication Date: 12 Aug 2005

Full Text: | Download PDF

Show Abstract
Most deaf children have parents who hear normally. Consequently most deaf adults who use American Sign Language (ASL) did not learn it from infancy from their parents. Rather, most deaf signers first learned to sign after infancy outside their homes. The experiment reported here examined the relationship between the length of time that signers have used sign language and their ability to repeat ASL utterances. Forty‐five, congenitally deaf, young adults both shadowed and recalled ASL sentences and scrambled‐ASL sentences (simultaneous reception and reproduction versus successive reception and reproduction). The signers were grouped according to the age at which they first learned to sign: 0, 5, 13–15, and after 18 years of age. The quantitative results show that the greater number of years a person has signed, the more accurately she or he can shadow and recall ASL utterances. More important, the qualitative results show that the groups make errors that are linguistically distinct: The first two groups perform like native users of spoken language, whereas the last two groups perform like non‐natives. The results demonstrate that fluency is a crucial factor in sign language perception. [Work supported by NTID and NINCDS.]
FREE

Vibrotactile perception of phonetic features of speech: A comparison of single‐channel and multi‐channel aids (A)

A. E. Carney

J. Acoust. Soc. Am. Volume 71, Issue S1, pp. S59-S59 (1982); (1 page)

Online Publication Date: 12 Aug 2005

Full Text: | Download PDF

Show Abstract
In an earlier paper [C. A. Beachler and A. E. Carney, J. Acoust. Soc. Am. Suppl. 1 69, S123 (1981)], two vibrotactile devices, one single‐channel and one multi‐channel, were compared with regard to their ability to transmit suprasegmental aspects of speech. Subjects trained with a single‐channel device were significantly better trained than those trained with a multi‐channel device in the identification of number of syllables in a word, syllable stress and intonation contour. As a follow up, two additional groups of subjects were trained with the same two different vibrotactile devices to recognize 20 consonants and eight vowels, all in a CV context. Subjects were trained in three conditions—visual only, vibrotactile only, and visual plus vibrotactile. Results indicated that there was essentially no difference between instruments in the percentage of phonemes transmitted correctly. Some slight differences between instruments were observed in the pattern of errors and the particular phonemes which were transmitted best by each instrument. These data are discussed with regard to the selective applicability of these devices to speech training for the hearing impaired.
FREE

Comparison of two tactile speech codes (A)

M. A. Clements, L. D. Braida, and N. I. Durlach

J. Acoust. Soc. Am. Volume 71, Issue S1, pp. S59-S59 (1982); (1 page)

Online Publication Date: 12 Aug 2005

Full Text: | Download PDF

Show Abstract
Two methods of encoding speech for tactile displays were compared in experiments which measured the discriminabilty of tactile representations of short speech segments. One display represented the short‐term speech spectrum in time‐swept mode and used vibration amplitude to encode spectral amplitude. The other represented the vocal‐tract shape derived from LPC analysis of the speech waveform and used the number of active vibrators to encode the logarithm of cross‐sectional area. Both displays were applied to the thigh via a 12 × 12 matrix of vibrators. The vibrators were driven at 250 Hz during voiced segments and by random noise during unvoiced segments. Overall results show a slight superiority for the spectral display in both vowel and consonant discrimination. Detailed results will be described in terms of an articulatory description of the speech stimuli, and a multidimensional scaling analysis of confusions. Theoretical analysis of the stimuli suggests that the detailed characteristics of the tactile patterns were only crudely discriminated. [Work supported by NSF.]
FREE

Vibrotactile communication of consonantal Information: Vowels mask consonants (A)

Barry G. Green

J. Acoust. Soc. Am. Volume 71, Issue S1, pp. S59-S59 (1982); (1 page)

Online Publication Date: 12 Aug 2005

Full Text: | Download PDF

Show Abstract
Identification of vibrotactile transformations of CVs was measured first for eight consonants (stops and nasals) in /_a/ context. The stimuli were 24‐channel spectral representations of the acoustic energy in the CVs, presented to the fingertip via the 6 × 24 matrix of an Optacon transducer. Performance reached only 50% to 60% correct, with nasals identified best, followed by voiceless and voiced stops. Confusions were greatest among stops that differed only in place of articulation. Perception of the place feature was examined in another experiment by testing identification of /b/, /d/, and /g/ with five different vowels. Place of articulation was increasingly well identified as the CVs were shortened in duration by truncating the vocalic portion of the stimulus: Performance stood at 54% correct when the CV lasted 258 ms, but rose to 77% correct when the CV lasted only 48 ms. This result indicates that vibrotactile communication of consonants may be poor because of temporal masking induced by the vocalic portion of the stimulus. [Supported by NSF.]
FREE

Evaluating a tactile aid on four‐year‐old profoundly deaf children (A)

Sandi L. Bond and Brian L. Scott

J. Acoust. Soc. Am. Volume 71, Issue S1, pp. S59-S60 (1982); (2 pages)

Online Publication Date: 12 Aug 2005

Full Text: | Download PDF

Show Abstract
Our purpose has been to develop a set of procedures for evaluating a speech reception aid on young, profoundly deaf children with limited linguistic skills. The subjects were two males and two female children ranging in age from 3 years and 7 months to 4 years and 3 months. Pure tone averages range from 93 dB HL to no response. The tasks reported are a subset of a broader range of tasks being investigated as tools for analyzing the effects of speech reception aids on the growth of linguistic skills. The current tasks have stressed focal training and reflect fundamental speech reception skills. At the most rudimentary level is the detection task, measuring the child's awareness of sound. The second task tests the child's ability to perceive and reproduce the syllable structure of brief utterances. The final task uses animal sounds to determine reception and production skills at a segmental level. The study ran for 6 weeks during which half the children began the study with the aid on and then switched to aid off after three weeks. The other two children began with the aid off, then switched to aid on. Results show a statistically significant difference in the rate of learning with the aid relative to the rate of learning without the aid. Data on each child will also be shown to illustrate the individual differences across subjects with differing linguistic skills. We believe this study illustrates a technique for fairly evaluating the efficacy of a sensory aid on young, prelingually deaf children. [Work supported by NINCDS.]
Return to All Sections
Close

Home Publications Meetings Contact ASA Site Map Back to Top

Copyright © Acoustical Society of America

close