Vowel and consonant recognition of cochlear implant patients using formant-estimating speech processors

Celebrating the one millionth cochlear implant

Cochlear implants have been the most successful neural prosthesis, with one million users globally. Researchers used the source-filter model and speech vocoder to design the modern multi-channel implants, allowing implantees to achieve 70%-80% correct sentence recognition in quiet, on average. Researchers also used the cochlear implant to help understand basic mechanisms underlying loudness, pitch, and cortical plasticity. While front-end processing advances improved speech recognition in noise, the unilateral implant speech recognition in quiet has plateaued since the early 1990s. This lack of progress calls for action on re-designing the cochlear stimulating interface and collaboration with the general neurotechnology community.

The role of glimpses with different energy in speech-in-noise recognition

Listeners can understand speech in noise by "glimpsing" some of the speech regions less affected by noise. This study investigates the contributions of those spectro-temporal regions, known as glimpses, at different energy levels to speech intelligibility in noise. Two listening experiments were conducted to examine the intelligibility of speech in different glimpse compositions in two types of noise. The results suggest that glimpsed spectro-temporal regions with energy above the mean noise level are the primary cue for speech perception in noise, and that listeners can use less-robust cues until at least 15 dB below the glimpsing threshold.

Validation of the Iowa Test of Consonant Perception

Speech perception (especially in background noise) is a critical problem for hearing-impaired listeners and an important issue for cognitive hearing science. Despite a plethora of standardized measures, few single-word closed-set tests uniformly sample the most frequently used phonemes and use response choices that equally sample phonetic features like place and voicing. The Iowa Test of Consonant Perception (ITCP) attempts to solve this. It is a proportionally balanced phonemic word recognition task designed to assess perception of the initial consonant of monosyllabic consonant-vowel-consonant (CVC) words. The ITCP consists of 120 sampled CVC words. Words were recorded from four different talkers (two female) and uniformly sampled from all four quadrants of the vowel space to control for coarticulation. Response choices on each trial are balanced to equate difficulty and sample a single phonetic feature. This study evaluated the psychometric properties of ITCP by examining reliability (test-retest) and validity in a sample of online normal-hearing participants. Ninety-eight participants completed two sessions of the ITCP along with standardized tests of words and sentence in noise (CNC words and AzBio sentences). The ITCP showed good test-retest reliability and convergent validity with two popular tests presented in noise. All the materials to use the ITCP or to construct your own version of the ITCP are freely available [Geller, McMurray, Holmes, and Choi (2020). https://osf.io/hycdu/].

[Phoneme discrimination training with experienced cochlear implant listeners]

BACKGROUND

Computer-based auditory training programmes seem to be a useful tool in the process of auditory rehabilitation after cochlear implantation (CI). Currently, little is known about the learning mechanism and efficiency of such programs. The aim of the study was to evaluate a specific auditory training programme for phoneme discrimination in experienced CI listeners.

MATERIALS AND METHODS

A total of 15 CI adult listeners with more than 2 years' CI experience participated in the auditory training. Over a period of 3 weeks they were instructed to train their phoneme discrimination via computer twice a week. Training material consisted of special syllables for consonants (vCv) and vowels (cVc) discrimination.

RESULTS

The discrimination abilities for consonants and vowels improved significantly over the training period for training group participants, whereas the changes for the consonants were higher. In addition, the improvement for voiced and unvoiced consonants was significant.

CONCLUSION

Computerised auditory training with phonemes improves CI listeners' discrimination abilities for consonants and vowels.

Predictors of audiological outcome following cochlear implantation in adults

The objective of this study was to examine variables that may predict open set speech discrimination following cochlear implantation. It consisted of a retrospective case review conducted in a tertiary referral centre with a cochlear implant programme. The patients were 117 postlingually deafened adult cochlear implant recipients. The main outcome measures were Bench, Kowal, Bamford (BKB) sentence scores recorded nine months following implant activation. The variables studied were age at the time of surgery, sex, duration of hearing loss, aetiology of hearing loss, residual hearing, implant type, speech processor strategy, number of active electrodes inserted. Variables found to have a significant effect on BKB following univariate analysis were entered into a multivariate analysis to determine independent predictors. Multivariate ordinal regression analysis gave an odds ration of 1.09 for each additional year of deafness prior to implantation (confidence interval 1.06-1.13; p < 0.001). Duration of deafness prior to implantation is an independent predictor of implant outcome. It accounted for 9% of the variability. Other factors must influence implant performance.

Fundamental frequency and speech intelligibility in background noise

Speech reception in noise is an especially difficult problem for listeners with hearing impairment as well as for users of cochlear implants (CIs). One likely cause of this is an inability to 'glimpse' a target talker in a fluctuating background, which has been linked to deficits in temporal fine-structure processing. A fine-structure cue that has the potential to be beneficial for speech reception in noise is fundamental frequency (F0). A challenging problem, however, is delivering the cue to these individuals. The benefits to speech intelligibility of F0 for both listeners with hearing impairment and users of CIs are reviewed, as well as various methods of delivering F0 to these listeners.

Perception and production of /r/ allophones improve with hearing from a cochlear implant

Tongue shape can vary greatly for allophones of /r/ produced in different phonetic contexts but the primary acoustic cue used by listeners, lowered F3, remains stable. For the current study, it was hypothesized that auditory feedback maintains the speech motor control mechanisms that are constraining acoustic variability of F3 in /r/; thus the listener's percept remains /r/ despite the range of articulatory configurations employed by the speaker. Given the potential importance of auditory feedback, postlingually deafened speakers should show larger acoustic variation in /r/ allophones than hearing controls, and auditory feedback from a cochlear implant could reduce that variation over time. To test these hypotheses, measures were made of phoneme perception and of production of tokens containing /r/, stop consonants, and /r/+stop clusters in hearing controls and in eight postlingually deafened adults pre- and postimplant. Postimplant, seven of the eight implant speakers did not differ from the control mean. It was also found that implant users' production of stop and stop+/r/ blend improved with time but the measured acoustic contrast between these was still better in the control speakers than for the implant group even after the implant users had experienced a year of improved auditory feedback.

Effects of short- and long-term changes in auditory feedback on vowel and sibilant contrasts

PURPOSE

To assess the effects of short- and long-term changes in auditory feedback on vowel and sibilant contrasts and to evaluate hypotheses arising from a model of speech motor planning.

METHOD

The perception and production of vowel and sibilant contrasts were measured in 8 postlingually deafened adults prior to activation of their cochlear implant speech processors, 1 month postactivation, and 1 year postactivation. Measures were taken postactivation both with and without auditory feedback. Contrast measures were also made for a group of speakers with reportedly normal hearing speaking with masked and unmasked auditory feedback.

RESULTS

Vowel and sibilant contrasts, measured in the absence of auditory feedback after 1 month of prosthesis use, were diminished compared with their values measured before prosthesis. Contrasts measured in the absence of auditory feedback after 1 year's experience with the prosthesis were increased compared with their values after 1 month's experience. In both time samples, contrasts were enhanced when auditory feedback was restored.

CONCLUSION

The provision of prosthetic hearing to postlingually deafened adults impaired their phonemic contrasts at first, as their auditory feedback had novel characteristics. Once auditory feedback became recalibrated with prosthesis use, it could, in turn, revise feedforward commands that control the contrasts in its absence.

Performance of patients using different cochlear implant systems: effects of input dynamic range

OBJECTIVE

To determine, for patients who had identical levels of performance on a monosyllabic word test presented in quiet, whether device differences would affect performance when tested with other materials and in other test conditions.

DESIGN

For Experiment 1, from a test population of 76 patients, three groups (N = 13 in each group) were created. Patients in the first group used the CII Bionic Ear behind-the-ear (BTE) speech processor, patients in the second group used the Esprit3G BTE speech processor, and patients in the third group used the Tempo+ BTE speech processor. The patients in each group were matched on (i) monosyllabic word scores in quiet, (ii) age at testing, (iii) duration of deafness, and (iv) experience with their device. Performance of the three groups was compared on a battery of tests of speech understanding, voice discrimination, and melody recognition. In Experiments 2 (N = 10) and 3 (N = 10) the effects of increasing input dynamic range in the 3G and CII devices, respectively, was assessed with sentence material presented at conversational levels in quiet, conversational levels in noise, and soft levels in quiet.

RESULTS

Experiment 1 revealed that patients fit with the CII processor achieved higher scores than Esprit3G and Tempo+ patients on tests of vowel recognition. CII and Tempo+ patients achieved higher scores than Esprit3G patients on difficult sentence material presented in noise at +10 and +5 dB SNR. CII patients achieved higher scores than Esprit3G patients on difficult sentence material presented at a soft level (54 dB SPL). Experiment 2 revealed that increasing input dynamic range in the Esprit3G device had (i) no effect at conversational levels in quiet, (ii) degraded performance in noise, and (iii) improved performance at soft levels. Experiment 3 revealed that increasing input dynamic range in the CII device improved performance in all conditions.

CONCLUSIONS

Differences in implant design can affect patient performance, especially in difficult listening situations. Input dynamic range and the method by which compression is implemented appear to be the major factors that account for our results.

Time course of speech changes in response to unanticipated short-term changes in hearing state

The timing of changes in parameters of speech production was investigated in six cochlear implant users by switching their implant microphones off and on a number of times in a single experimental session. The subjects repeated four short, two-word utterances, /dV1n#SV2d/ (S = /s/ or /S/), in quasi-random order. The changes between hearing and nonhearing states were introduced by a voice-activated switch at V1 onset. "Postural" measures were made of vowel sound pressure level (SPL), duration, F0; contrast measures were made of vowel separation (distance between pair members in the formant plane) and sibilant separation (difference in spectral means). Changes in parameter values were averaged over multiple utterances, lined up with respect to the switch. No matter whether prosthetic hearing was blocked or restored, contrast measures for vowels and sibilants did not change systematically. Some changes in duration, SPL and F0 were observed during the vowel within which hearing state was changed, V1, as well as during V2 and subsequent utterance repetitions. Thus, sound segment contrasts appear to be controlled differently from the postural parameters of speaking rate and average SPL and F0. These findings are interpreted in terms of the function of hypothesized feedback and feedforward mechanisms for speech motor control.

On the structure of phoneme categories in listeners with cochlear implants

PURPOSE

To describe cochlear implant users' phoneme labeling, discrimination, and prototypes for a vowel and a sibilant contrast, and to assess the effects of 1 year's experience with prosthetic hearing.

METHOD

Based on naturally produced clear examples of "boot," "beet," "said," and "shed" by 1 male and 1 female speaker, continua with 13 stimuli were synthesized for each contrast. Seven hearing controls labeled those stimuli and assigned them goodness ratings, as did 7 implant users at 1-month postimplant. One year later, these measures were repeated, and within category discrimination, d', was assessed.

RESULTS

Compared with controls, implant users' vowel and sibilant labeling slopes were substantially shallower but improved over 1 year of prosthesis use. Their sensitivity to phonetic differences within phoneme categories was about half that of controls. The slopes of their goodness rating functions were shallower and did not improve. Their prototypes for the sibilant contrast (but not the vowels) were closer to one another and did not improve by moving apart.

CONCLUSIONS

Implant users' phoneme labeling and within-category perceptual structure were anomalous at 1-month postimplant. After 1 year of prosthesis use, phoneme labeling categories had sharpened but within category discrimination was well below that of hearing controls.

Effects of masking noise on vowel and sibilant contrasts in normal-hearing speakers and postlingually deafened cochlear implant users

The role of auditory feedback in speech production was investigated by examining speakers' phonemic contrasts produced under increases in the noise to signal ratio (N/S). Seven cochlear implant users and seven normal-hearing controls pronounced utterances containing the vowels /i/, /u/, /e/ and /ae/ and the sibilants /s/ and /I/ while hearing their speech mixed with noise at seven equally spaced levels between their thresholds of detection and discomfort. Speakers' average vowel duration and SPL generally rose with increasing N/S. Average vowel contrast was initially flat or rising; at higher N/S levels, it fell. A contrast increase is interpreted as reflecting speakers' attempts to maintain clarity under degraded acoustic transmission conditions. As N/S increased, speakers could detect the extent of their phonemic contrasts less effectively, and the competing influence of economy of effort led to contrast decrements. The sibilant contrast was more vulnerable to noise; it decreased over the entire range of increasing N/S for controls and was variable for implant users. The results are interpreted as reflecting the combined influences of a clarity constraint, economy of effort and the effect of masking on achieving auditory phonemic goals-with implant users less able to increase contrasts in noise than controls.

Effects of bite blocks and hearing status on vowel production

This study explores the effects of hearing status and bite blocks on vowel production. Normal-hearing controls and postlingually deaf adults read elicitation lists of /hVd/ syllables with and without bite blocks and auditory feedback. Deaf participants' auditory feedback was provided by a cochlear prosthesis and interrupted by switching off their implant microphones. Recording sessions were held before prosthesis was provided and one month and one year after. Long-term absence of auditory feedback was associated with heightened dispersion of vowel tokens, which was inflated further by inserting bite blocks. The restoration of some hearing with prosthesis reduced dispersion. Deaf speakers' vowel spaces were reduced in size compared to controls. Insertion of bite blocks reduced them further because of the speakers' incomplete compensation. A year of prosthesis use increased vowel contrast with feedback during elicitation. These findings support the inference that models of speech production must assign a role to auditory feedback in error-based correction of feedforward commands for subsequent articulatory gestures.

Phoneme recognition and confusions with multichannel cochlear implants: vowels

The aim of this study was to investigate how postlingually severely or profoundly hearing-impaired adults relearn to recognize vowels after receiving multichannel cochlear implants. Vowel recognition of 19 Finnish-speaking subjects was studied for a minimum of 6 months and a maximum of 24 months using an open-set nonsense-syllable test in a prospective repeated-measure design. The responses were coded for phoneme errors, and 95% confidence intervals for recognition and confusions were calculated. The average vowel recognition was 68% (95% confidence interval = 66-70%) 6 months after switch-on and 80% (95% confidence interval = 78-82%) 24 months after switch-on. The vowels [ae], [u], [i], [o], and [a] were the easiest to recognize, and the vowels [y], [e], and [ø] were the most difficult. In conclusion, adaptation to electrical hearing using a multichannel cochlear implant was achieved well; but for at least 2 years, given two vowels with either F1 or F2 at roughly the some frequencies, confusions were drawn more towards the closest vowel with the next highest F1 or F2.

Phoneme recognition and confusions with multichannel cochlear implants: consonants

The aim of this study was to investigate how postlingually severely or profoundly hearing-impaired adults relearn to recognize consonants after receiving multichannel cochlear implants. Consonant recognition of 19 Finnish-speaking subjects was studied for a minimum of 6 months and a maximum of 24 months using an open-set nonsense-syllable test in a prospective repeated-measure design. Responses were coded for phoneme errors, and proportions of correct responses and 95% confidence intervals were calculated for recognition and confusions. Two years after the switch-on, the mean recognition of consonants was 71% (95% confidence interval = 68-73%). The manner of articulation was easier to classify than the place of articulation, and the consonants [s], [r], [k], [t], [p], [n], and [j] were easier to recognize than [h], [m], [l], and [v]. Adaptation to electrical hearing with a multichannel cochlear implant was successful, but consonants with alveolar, palatal, or velar transitions (high F2) were better recognized than consonants with labial transitions (low F2). The locus of the F2 transitions of the consonants with better recognition was at the frequencies 1.5-2 kHz, whereas the locus of the F2 transitions of the consonants with poorer recognition was at 1.2-1.4 kHz. A tendency to confuse consonants with the closest consonant with higher F2 transition was also noted.

Contribution of speech rate to speech perception in multichannel cochlear implant users

This study describes the effect of speech rate (fast, 11 syllables per second; medium, 9 syllables per second; slow, 6 syllables per second) on speech perception in 10 cochlear implant users. The speech perception performance was evaluated on the basis of the percentage score of syllables that were correctly recalled in sentences composed of 4 to 6 words. The percentage scores at the fast, medium, and slow speech rates were 15.7%, 39.0%, and 56.0%, respectively. The effect of speech rate slowing was significant (p < .0001). Variations in the effect of speech rate slowing were observed in the cochlear implant users. The improvement of speech perception by speech rate slowing was significantly (p < .005) related to the word test score and the score at the fast speech rate. The results reveal that the rate of speech is an important factor in improving the speech perception of cochlear implant users.

Use of vocalic information in the identification of /s/ and /sh/ by children with cochlear implants

OBJECTIVE

When a syllable such as "sea" or "she" is spoken, listeners with normal hearing extract evidence of the fricative consonant from both the fricative noise and the following vocalic segment. If the fricative noise is made ambiguous, listeners may still perceive "s" or "sh" categorically, depending on information in the vocalic segment. Do children whose auditory experience comes from electrical stimulation also display this effect, in which a subsequent segment of speech disambiguates an earlier segment?

DESIGN

Unambiguous vowels were appended to ambiguous fricative noises to form tokens of the words "she," "sea," "shoe," and "Sue." A four-choice identification test was undertaken by children with normal hearing (N = 29), prelingually deaf children with the Nucleus Spectra-22 implant system using the SPEAK coding strategy (N = 13), postlingually deafened adults with the same implant system (N = 26), and adults with normal hearing (N = 10). The last group undertook the test before and after the stimuli were processed to simulate the transformations introduced by the SPEAK coding strategy.

RESULTS

All four groups made use of vocalic information. Simulated processing reduced the use made by normal-hearing adults. Implanted subjects made less use than the other groups, with no significant difference between implanted children and implanted adults. The highest levels of use by implanted subjects were within one standard deviation of the mean level displayed when normal-hearing adults listened to processed stimuli. Analyses showed that the SPEAK strategy distorted formant contours in the vocalic segments of the stimuli in ways that are compatible with the errors of identification made by implanted subjects.

CONCLUSIONS

Some children with implants can extract information from a following vowel to disambiguate a preceding fricative noise. The upper limit on this ability may be set by distortions introduced by the implant processor, rather than by the auditory experience of the child.

The influence of contextual information on the perception of speech by postlingually and prelingually profoundly hearing-impaired Hebrew-speaking adolescents and adults

OBJECTIVE

To study linguistic context effects on speech perception by prelingually (N = 10) and postlingually (N = 10) hearing-impaired Hebrew-speaking adolescents and adults.

DESIGN

Participants were cochlear implant surgery candidates with profound sensorineural hearing loss (95 dB HL+). Four levels of speech perception materials were presented through the auditory-visual modality: monosyllabic nonsense syllables (phonological content), monosyllabic meaningful words (semantic-lexical), sentences without presented topics (semantic-syntactic), and sentences with topics (topical). Data were analyzed via percentage correct recognition and via k andj factors (Boothroyd & Nittrouer, 1988).

RESULTS

Both participant groups performed better when speech material having more contextual information was presented, with no significant intergroup differences.

CONCLUSIONS

This similar usage of contextual information among prelingually and postlingually hearing-impaired participants is encouraging and suggests the role of factors other than age of onset.

Fos-like immunoreactivity in the auditory brainstem evoked by bipolar intracochlear electrical stimulation: effects of current level and pulse duration

Fos-like immunoreactivity was used to compare the auditory brain stem excitation elicited by bipolar electrical stimulation of the cochlea at various current levels relative to the electrically evoked auditory brain stem response threshold for a 50-micros/phase monophasic pulse. Fos-like immunoreactive cells were labeled in primary auditory brain stem regions. The distribution of labeled cells was restricted to regions known to be cochleotopically related to the stimulated region of the scala tympani. Some labeled cells were observed at 2x electrically evoked auditory brain stem response threshold. The number, density and spatial distribution of labeled cells were quantified in the dorsal cochlear nucleus and inferior colliculus, and found to increase with increasing level of stimulation. For 50-micros pulses, the location of labeled neurons remained reasonably restricted to narrow bands within each region until the 1Ox level of stimulation (20 dB above electrically evoked auditory brain stem response threshold) was reached. While a monotonic increase in Fos-like immunoreactivity with increasing stimulus level was observed in most nuclei, for cells of the superficial layer of the dorsal cochlear nucleus, a non-monotonic change with increasing stimulus level was seen. This dorsal cochlear nucleus non-monotonicity may indicate that, at higher levels of stimulation, a secondary indirect inhibitory input, probably associated with activation of deep layer dorsal cochlear nucleus cells, reduces excitatory responses at the superficial layer of the dorsal cochlear nucleus. Electrically evoked auditory brain stem response and Fos expression showed parallel changes as a function of stimulus level and pulse duration. The data indicate that discrete activation of cell populations within the central auditory pathways can occur with bipolar electrical stimulation to the highest levels of stimulation typically useful in humans. The data also indicate a close, but not identical, quantitative relationship between Fos-like immunoreactivity and electrophysiological response amplitude. These findings support the view that a study of Fos-like immunoreactivity can provide a powerful and quantitative tool for study of the dynamic response characteristics of cells of the central auditory system to electrical stimulation at suprathreshold levels. The data suggest that there is a monotonic increase in the number of neurons responsive to intracochlear electrical stimulation as a function of stimulus level, at least through the upper half of the dynamic range, but that this increase does not result in a complete loss of spatial selectivity. Coupled with previous psychophysical studies, these results suggest that the increase in the number of activated neurons is functionally beneficial, resulting in improved discrimination of changes in the electrical signals.

Natural vowel and consonant recognition by Laura cochlear implantees

OBJECTIVE

This study was performed to examine whether spectral and temporal properties of speech sounds are transferred effectively by the Laura cochlear implant device, and to determine whether listeners with different levels of performance use the same type of information to understand speech.

DESIGN

Twenty-five adult Laura cochlear implantees identified 12 repetitions of 10 vowels in /hVt/ utterances and 16 consonants placed in intervocalic /a/ context. The stimulus-response confusion matrices were analyzed in terms of relative information transmission scores (Miller & Nicely, 1955), and they were subjected to multidimensional INDSCAL analyses to examine whether a similar pattern of results would emerge when no prior assumptions were made about the division of categories.

RESULTS

To examine perception for different levels of performance, the vowel and consonant confusion data were divided into three groups of subjects: the better, intermediate, and poorer performers. In general, the INDSCAL analyses confirmed the results obtained from the information transmission analyses. However, they also supplemented it with other perceptually relevant cues for cochlear implantees, as well as with the weightings over different perceptual dimensions for different types of performers. These analyses suggest that although all subjects use the same type of information, the better performers are more capable of using these different cues than poor performers.

Investigation of the effects of temporal and spatial interactions on speech-recognition skills in cochlear-implant subjects

Forward masking was investigated as a measure of spectral and temporal interactions. Such interactions may adversely affect speech recognition in cochlear-implant subjects. Seven subjects, implanted with the Nucleus 22 device, performed a forward-masking task. They also performed an electrode-discrimination task in order to measure spectral interactions without temporal interactions. Correlation analysis indicated a significant relationship between data obtained in the two tasks (p < 0.1). The two tasks were also correlated with the subjects' scores from five measures of speech recognition. Forward masking and electrode discrimination were strongly correlated with measures requiring consonant and phoneme recognition, respectively. These results indicate that the relationship between forward masking and speech recognition may be due, in part, to a lack of spectral resolution. The data also indicate that consonants may be more readily masked than vowels. Forward-masking data measured for all clinically programmed electrodes in three of the seven subjects were used with a model of the spectral maxima sound processor (SMSP) to estimate the number of electrodes stimulated during a consonant that might be masked by prior presentation of a vowel. These results suggest that temporal interactions across electrodes may be a factor in speech-recognition abilities of some cochlear-implant subjects.

A 4-year investigation into phonetic inventory development in young cochlear implant users

Phonetic inventories of 9 children with profoundly impaired hearing who used the 22-electrode cochlear implant (Cochlear Limited) were monitored before implantation and during the first 4 years of implant use. All children were 5 years old or younger at the time of implant. Spontaneous speech samples were collected at regular intervals for each child and analyzed to investigate phone acquisition over the post-implant period. Acquisition was measured using two different criteria. The "targetless" criterion required the child to produce a phonetically recognizable sound spontaneously, and the "target" criterion required the child to produce the phone correctly at least 50% of the time in meaningful words. At 4 years post-implant, 40 out of 44 phones (91 %) had reached the targetless criterion, and 29 phones (66%) had reached the target criterion for 5 or more of the children. Over the time of the study 100% of monophthongs, 63% of diphthongs, and 54% of consonants reached the target criterion. The average time taken for a phone to progress from the targetless to target criterion was 15 months. Overall, the data suggest trends in the order of phone acquisition similar to those of normally hearing children, although the process of acquisition occurred at a slower rate.

Aided speech recognition abilities of adults with a severe or severe-to-profound hearing loss

Adults with severe or severe-to-profound hearing losses constitute between 11% and 13.5% of the hearing-impaired population. A detailed investigation of the speech recognition of adults with severe (n = 20) or severe-to-profound (n = 14) hearing loss was conducted at The University of Melbourne. Each participant took part in a series of speech recognition tasks while wearing his or her currently fitted hearing aid(s). The assessments included closed-set tests of consonant recognition and vowel recognition, combined with open-set tests of monosyllabic word recognition and sentence recognition. Sentences were presented in quiet and in noise at +10 dB SNR to replicate an environment more typical of everyday listening conditions. Although the results demonstrated wide variability in performance, some general trends were observed. As expected vowels were generally well perceived compared with consonants. Monosyllabic word recognition scores for both the adults with a severe hearing impairment (M = 67.2%) and the adults with a severe-to-profound hearing impairment (M = 38.6%) could be predicted from the segmental tests, with an allowance for lexical effects. Scores for sentences presented in quiet showed additional linguistic effects and a significant decrease in performance with the addition of background noise (from 82.9% to 74.1% for adults with a severe hearing loss and from 55.8% to 34.2% for adults with a severe-to-profound hearing loss). Comparisons were made between the participants and a group of adults using a multiple-channel cochlear implant. This comparison indicated that some adults with a severe or severe-to-profound hearing loss may benefit from the use of a cochlear implant. The results of this study support the contention that cochlear implant candidacy should not rely solely on audiometric thresholds.

The recognition of vowels produced by men, women, boys, and girls by cochlear implant patients using a six-channel CIS processor

Five patients who used a six-channel, continuous interleaved sampling (CIS) cochlear implant were presented vowels, in two experiments, from a large sample of men, women, boys, and girls for identification. At issue in the first experiment was whether vowels from one speaker group, i.e., men, were more identifiable than vowels from other speaker groups. At issue in the second experiment was the role of the fifth and sixth channels in the identification of vowels from the different speaker groups. It was found in experiment 1 that (i) the vowels produced by men were easier to identify than vowels produced by any of the other speaker groups, (ii) vowels from women and boys were more difficult to identify than vowels from men but less difficult than vowels from girls, and (iii) vowels from girls were more difficult to identify than vowels from all other groups. In experiment 2 removal of channels 5 and 6 from the processor impaired the identification of vowels produced by women, boys and girls but did not impair the identification of vowels produced by men. The results of experiment 1 demonstrate that scores on tests of vowels produced by men overestimate the ability of patients to recognize vowels in the broader context of multi-talker communication. The results of experiment 2 demonstrate that channels 5 and 6 become more important for vowel recognition as the second formants of the speakers increase in frequency.

Effect of relative amplitude and formant transitions on perception of place of articulation by adult listeners with cochlear implants

Previous studies have shown that manipulation of a particular frequency region of the consonantal portion of a syllable relative to the amplitude of the same frequency region in an adjacent vowel influences the perception of place of articulation. This manipulation has been called the relative amplitude cue. Earlier studies have examined the effect of relative amplitude and formant transition manipulations upon labeling place of articulation for fricatives and stop consonants in listeners with normal hearing. The current study sought to determine if (a) the relative amplitude cue is used by adult listeners wearing a cochlear implant to label place of articulation, and (b) adult listeners wearing a cochlear implant integrated the relative amplitude and formant transition information differently than listeners with normal hearing. Sixteen listeners participated in the study, 12 with normal hearing and 4 postlingually deafened adults wearing the Nucleus 22 electrode Mini Speech Processor implant with the multipeak processing strategy. The stimuli used were synthetic consonant-vowel (CV) syllables in which relative amplitude and formant transitions were manipulated. The two speech contrasts examined were the voiceless fricative contrast /s/-"sh" and the voiceless stop consonant contrast /p/-/t/. For each contrast, listeners were asked to label the consonant sound in the syllable from the two response alternatives. Results showed that (a) listeners wearing this implant could use relative amplitude to consistently label place of articulation, and (b) listeners with normal hearing integrated the relative amplitude and formant transition information more than listeners wearing a cochlear implant, who weighted the relative amplitude information as much as 13 times that of the transition information.

Changes in synthetic and natural vowel perception after specific training for congenitally deafened patients using a multichannel cochlear implant

OBJECTIVE

The aim was to determine whether the ability to use place-coded vowel formant information could be improved after training in a group of congenitally deafened patients, who showed limited speech perception ability after cochlear implant use ranging from 1 yr 8 mo to 6 yr 11 mo. A further aim was to investigate the relationship between electrode position difference limens and vowel recognition.

DESIGN

Three children, one adolescent, and one young adult were assessed with synthesized versions of the words/hid, head, had, hud, hod, hood/containing three formants and with a natural version of these words as well as with a 12-alternative, closed-set task containing monosyllabic words. The change in performance during a nontraining period was compared to the change in performance after 10 training sessions.

RESULTS

After training, two children showed significant gains on a number of tests and improvements were consistent with their electrode discrimination ability. Difference limens ranged from one to three electrodes for these patients as well as for two other patients who showed minimal to no improvements. The minimal gains shown by the final patient could be partly explained by poorer apical electrode position difference limen.

CONCLUSIONS

Significant gains in vowel perception occurred post-training on several assessments for two of the children. This suggests the need for children to continue to have aural rehabilitation for a substantial period after implantation. Minimal improvements, however, occurred for the remaining patients. With the exception of one patient, their poorer performance was not associated with poorer electrode discrimination.

Psychophysical studies with two binaural cochlear implant subjects

Psychophysical studies have been completed with two binaural cochlear implant patients. In our earlier studies [van Hoesel et al., J. Acoust. Soc. Am. 94, 3187-3189 (1993); R. J. M. van Hoesel and G. M. Clark, Ann. Otol. Rhinol. Laryngol. Suppl. 106 104, 233-235 (1995)], lateralization experiments showed good sensitivity to interaural amplitudes but poor sensitivity to interaural time delays when compared with normal hearing subjects. In the studies presented here, both temporal and binaural intensity interactions were further explored. Interaural time delay (ITD) perception was investigated using direct measurement of the just-noticeable difference (jnd) in ITD. Both rate and place of stimulation were varied. Binaural rate discrimination was measured and compared with monaural rate perception. Binaural intensity interaction was explored for matched and unmatched place conditions by means of loudness summation and central masking studies. Results showed that ITDs for interaural time delays were large when compared to normal hearing, even when place of stimulation on each of the two sides was carefully matched. The jnds in ITD were similar for stimulation rates from 50 to 200 pps, and increased at 300 pps. Rate difference limens experiments showed similar results for diotic and monaural stimuli, but improved jnds for dichotic presentation at stimulation rates below 150-200 pps. Binaural intensity interactions showed loudness summation effects with both patients, for matched as well as unmatched place conditions. Central masking was also observed with both subjects, although it was not found to be place dependent.

Some considerations in evaluating spoken word recognition by normal-hearing, noise-masked normal-hearing, and cochlear implant listeners. I: The effects of response format

OBJECTIVE

The purpose of the present studies was to assess the validity of using closed-set response formats to measure two cognitive processes essential for recognizing spoken words---perceptual normalization (the ability to accommodate acoustic-phonetic variability) and lexical discrimination (the ability to isolate words in the mental lexicon). In addition, the experiments were designed to examine the effects of response format on evaluation of these two abilities in normal-hearing (NH), noise-masked normal-hearing (NMNH), and cochlear implant (CI) subject populations.

DESIGN

The speech recognition performance of NH, NMNH, and CI listeners was measured using both open- and closed-set response formats under a number of experimental conditions. To assess talker normalization abilities, identification scores for words produced by a single talker were compared with recognition performance for items produced by multiple talkers. To examine lexical discrimination, performance for words that are phonetically similar to many other words (hard words) was compared with scores for items with few phonetically similar competitors (easy words).

RESULTS

Open-set word identification for all subjects was significantly poorer when stimuli were produced in lists with multiple talkers compared with conditions in which all of the words were spoken by a single talker. Open-set word recognition also was better for lexically easy compared with lexically hard words. Closed-set tests, in contrast, failed to reveal the effects of either talker variability or lexical difficulty even when the response alternatives provided were systematically selected to maximize confusability with target items.

CONCLUSIONS

These findings suggest that, although closed-set tests may provide important information for clinical assessment of speech perception, they may not adequately evaluate a number of cognitive processes that are necessary for recognizing spoken words. The parallel results obtained across all subject groups indicate that NH, NMNH, and CI listeners engage similar perceptual operations to identify spoken words. Implications of these findings for the design of new test batteries that can provide comprehensive evaluations of the individual capacities needed for processing spoken language are discussed.

Evaluation of the Nucleus Spectra 22 processor and new speech processing strategy (SPEAK) in postlinguistically deafened adults

A new speech processing strategy (SPEAK) has been compared with the previous Multipeak (MPEAK) strategy in a study with 24 postlinguistically deafened adults. The results show that performance with the SPEAK coding strategy was significantly better for 58.31% of subjects on closed-set consonant identification, for 33.3% of subjects on closed-set vowel identification and open-set monosyllabic word recognition, and for 81.8% of subjects on open-set sentence recognition in quiet and in competing noise (+ 10 dB signal-to-noise ratio). By far the largest improvement observed was for sentence recognition in noise, with the mean score across subjects for the SPEAK strategy twice that obtained with MPEAK.

Performance of adult Ineraid and Nucleus cochlear implant patients after 3.5 years of use

Forty-two postlingually deafened adult patients, 21 with a formant extraction version of the Nucleus cochlear implant and 21 with the Ineraid cochlear implant (analog processing), were evaluated on a series of speech perception tests after using their implants for about 3.5 years. A wide range of performance was observed across patients for both devices. All but 4 patients showed an enhancement in their lipreading ability with the implant. Word recognition averaged about 14-19% correct, and word recognition in sentences averaged about 43-49% correct for the two implant groups. Average performance was superior with the Ineraid implant on consonant recognition in noise. An information transmission analysis suggested that vowel perception was influenced by first- and third-formant frequency for the Nucleus, and first-formant and fundamental frequency for the Ineraid patients. It appeared that the Ineraid device was more effective, on average, at conveying information about consonant nasality and frication. For consonant perception, nasality and frication contributed most to the total information transmitted for both implant types. Both devices had difficulty conveying information about vowel second-formant frequency and consonant place information. These scores at 3.5 years are substantially elevated from preoperative performance and, overall, the patients clearly benefit from their implant.

Vowel and diphthong production by young users of cochlear implants and the relationship between the phonetic level evaluation and spontaneous speech

This investigation assessed how the vowel and diphthong production of young users of cochlear implants varied over time and how performance on the Phonetic Level Evaluation (PLE, Ling, 1976) corresponded with vowel and diphthong production during spontaneous speech. Eight children with prelingual deafness were tested with the PLE on five occasions: before receiving a Cochlear Corporation Nucleus cochlear implant and at the following points after receiving a cochlear implant: 6 months, 12 months, 18 months, and 24 or 36 months. An audiovideo recording of spontaneous speech was obtained at each test session. Performance on the PLE was scored with the scoring system designed by Kirk and Hill-Brown (1985). Performance during spontaneous speaking was scored by referencing a transcription of the spoken message to a transcription of the signed message. The correlations between the PLE and the spontaneous speech measures were weak, suggesting that performance on the PLE has low predictive value for vowel and diphthong production during spontaneous speaking. The results from the spontaneous speech samples collected over time suggest that two changes occurred: (a) vowel and diphthong production became more diverse and (b) production became more accurate. It is suggested that increased access to formant information enables subjects to enlarge their system of phonological performance and refine their motoric ability to establish vowel and diphthong targets.

A comparison of speech perception of cochlear implantees using the Spectral Maxima Sound Processor (SMSP) and the MSP (MULTIPEAK) processor

The Spectral Maxima Sound Processor (SMSP) is a portable speech processor which has recently been developed at the University of Melbourne for use with multiple-electrode cochlear implants. In this processor, the six largest outputs (maxima) of 16 bandpass filters are used to stimulate the cochlea on a place basis at a constant rate. This speech processing strategy has been compared with the MSP(MULTIPEAK) strategy, in which four electrodes are selected for stimulation in every glottal pulse period. The study was undertaken on four postlinguistically deaf adults. The results show that, for this group of subjects, the performance of the SMSP processor was significantly better than that of the MSP(MULTIPEAK) processor for the recognition of closed-set vowels and consonants, open-set monosyllabic words, and open-set sentences in noise, when using electrical stimulation alone. The SMSP mean scores were: vowels 91.3%, consonants 74.9%, words 57.4%, and sentences in noise 78.7%. The MSP(MULTIPEAK) mean scores were: vowels 76.3%, consonants 59.4%, words 39.9%, and sentences in noise 50.0%.

Factors predicting postoperative sentence scores in postlinguistically deaf adult cochlear implant patients

A sample of 64 postlinguistically profoundly to totally deaf adult cochlear implant patients were tested without lipreading by means of the Central Institute for the Deaf (CID) sentence test 3 months postoperatively. Preoperative promontory stimulation results (thresholds, gap detection, and frequency discrimination), age, duration of profound deafness, cause of deafness, lipreading ability, postoperative intracochlear thresholds and dynamic ranges for electrical stimulation, depth of insertion of the electrode array into the scala tympani, and number of electrodes in use were considered as possible factors that might be related to the postoperative sentence scores. A multiple regression analysis with stepwise inclusion of independent variables indicated that good gap detection and frequency discrimination during preoperative promontory testing, larger numbers of electrodes in use, and greater dynamic ranges for intracochlear electrical stimulation were associated with better CID scores. The CID scores tended to decrease with longer periods of profound deafness.

Psychophysical studies using a multiple-electrode cochlear implant in patients who were deafened early in life

Psychophysical studies were conducted on 10 cochlear implant patients, between 5 and 23 years of age at the time of surgery, who were deafened prior to 4 years of age. The multiple-electrode prosthesis manufactured by Cochlear Ltd. was used. Identification studies, the recognition of 2-4 stimuli after some training, were conducted on 3 of the 10 patients. For current level and repetition rate identification, performance was comparable to that observed for postlingual adult patients. For electrode position identification, however, performance was much poorer than that observed for postlingual adults. In general, the difference limens for current level, repetition rate and duration, and the gap detection thresholds were similar to those observed for postlingual adults. For 3 patients whose etiology was Usher's syndrome, the repetition rate limens at higher rates were larger than those of the other patients. The limens for electrode position, measured in a discrimination task, were 1-3 electrodes for most patients. However, for 3 patients, limens of 6-10 electrodes were recorded. For numerosity judgements, the counting of stimuli in a temporal series as a function of the rate of presentation, the patients were less successful at counting for rates of 3-8/s than for lower rates (1-2/s).

Natural vowel perception by patients with the ineraid cochlear implant

Vowel recognition was tested in 10 patients using the Ineraid cochlear implant. The vowels were produced by a male speaker in the context 'heed, hid, head, had, hawed, hood, who'd, hud' and 'heard'. Performance varied from 34 to 93% correct. A descriptive feature system for the vowels was determined from an acoustic analysis. An information transfer analysis of these features suggested that information about the first formant frequency, vowel duration and fundamental frequency was transmitted. Information about the second and third formant frequency was transmitted less well. A sequential information transmission analysis suggested that the features of the first formant and duration accounted for nearly 80% of the information transmitted. The fundamental frequency and second formant frequency information accounted for an additional 8%. Information provided by the third formant frequency was largely redundant.

An advanced multiple channel cochlear implant

An advanced multiple channel cochlear implant hearing prosthesis is described. Stimulation is presented through an array of 20 electrodes located in the scala tympani. Any two electrodes can be configured as a bipolar pair to conduct a symmetrical, biphasic, constant-current pulsatile stimulus. Up to three stimuli can be presented in rapid succession or effectively simultaneously. For simultaneous stimulation, a novel time-division current multiplexing technique has been developed to obviate electrode interactions that may compromise safety. The stimuli are independently controllable in current amplitude, duration, and onset time. Groups of three stimuli can be generated at a rate of typically 500 Hz. Stimulus control data and power are conveyed to the implant through a single transcutaneous inductive link. The device also incorporates a telemetry system that enables electrode voltage waveforms to be monitored externally in real time. The electronics of the implant are contained almost entirely on a custom designed integrated circuit. Preliminary results obtained with the first patient to receive the advanced implant are included.

Synthetic two-formant vowel perception by some of the better cochlear-implant patients

Synthetic two-formant vowel recognition was tested in some of the better cochlear-implant users to determine their ability to resolve sounds that differ essentially in their frequency content. The vowels /i, epsilon, alpha, u/ were synthesized with equal duration and similar sound pressure level. The formant values were chosen to approximate the values of these vowels in French, German and American English. Performance ranged from 29 to 71% for the 6 patients with the Chorimac implant, from 25 to 67% for the 9 patients with the 3M/Vienna implant, from 63 to 92% for the 10 patients with the Nucleus implant from Hannover, from 17 to 79% for the 10 patients with the Duren/Cologne implant, from 54 to 100% for the 9 patients with the Symbion implant, and from 79 to 100% for the 10 Nucleus patients from the USA. Patients with each of these devices can utilize some spectral information when recognizing steady-state vowels.