Lexical tone recognition in noise in normal-hearing children and prelingually deafened children with cochlear implants

Lexical tone recognition in multi-talker babbles and steady-state noise by Mandarin-speaking children with unilateral cochlear implants or bimodal hearing

BACKGROUND AND OBJECTIVES

Lexical tone presents challenges to cochlear implant (CI) users especially in noise conditions. Bimodal hearing utilizes residual acoustic hearing in the contralateral side and may offer benefits for tone recognition in noise. The purpose of the present study was to evaluate tone recognition in both steady-state noise and multi-talker babbles by the prelingually-deafened, Mandarin-speaking children with unilateral CIs or bimodal hearing.

METHODS

Fifty-three prelingually-deafened, Mandarin-speaking children who received CIs participated in this study. Twenty-two of them were unilateral CI users and 31 wore a hearing aid (HA) in the contralateral ear (i.e., bimodal hearing). All subjects were tested for Mandarin tone recognition in quiet and in two types of maskers: speech-spectrum-shaped noise (SSN) and two-talker babbles (TTB) at four signal-to-noise ratios (-6, 0, +6, and +12 dB).

RESULTS

While no differences existed in tone recognition in quiet between the two groups, the Bimodal group outperformed the Unilateral CI group under noise conditions. The differences between the two groups were significant at SNRs of 0, +6, and +12 dB in the SSN conditions (all p < 0.05), and at SNRs of +6 and +12 dB of TTB conditions (both p < 0.01), but not significant at other conditions (p > 0.05). The TTB exerted a greater masking effect than the SSN for tone recognition in the Unilateral CI group as well as in the Bimodal group at all SNRs tested (all p < 0.05). Among demographic or audiometric variables, only age at implantation showed a weak but significant correlation with the mean tone recognition performance under the SSN conditions (r = -0.276, p = 0.045). However, when Bonferroni correction was applied to the correlation analysis results, the weak correlation became not significant.

CONCLUSION

Prelingually-deafened children with CIs face challenges in tone perception in noisy environments, especially when the noise is fluctuating in amplitude such as the multi-talker babbles. Wearing a HA on the contralateral side when residual hearing permits is beneficial for tone recognition in noise.

Investigating Perception to Production Transfer in Children With Cochlear Implants: A High Variability Phonetic Training Study

PURPOSE

This study builds upon an established effective training method to investigate the advantages of high variability phonetic identification training for enhancing lexical tone perception and production in Mandarin-speaking pediatric cochlear implant (CI) recipients, who typically face ongoing challenges in these areas.

METHOD

Thirty-two Mandarin-speaking children with CIs were quasirandomly assigned into the training group (TG) and the control group (CG). The 16 TG participants received five sessions of high variability phonetic training (HVPT) within a period of 3 weeks. The CG participants did not receive the training. Perception and production of Mandarin tones were administered before (pretest) and immediately after (posttest) the completion of HVPT via lexical tone recognition task and picture naming task. Both groups participated in the identical pretest and posttest with the same time frame between the two test sessions.

RESULTS

TG showed significant improvement from pretest to posttest in identifying Mandarin tones for both trained and untrained speech stimuli. Moreover, perceptual learning of HVPT significantly facilitated trainees' production of T1 and T2 as rated by a cohort of 10 Mandarin-speaking adults with normal hearing, which was corroborated by acoustic analyses revealing improved fundamental frequency (F0) median for T1 and T2 production and enlarged F0 movement for T2 production. In contrast, TG children's production of T3 and T4 showed nonsignificant changes across two test sessions. Meanwhile, CG did not exhibit significant changes in either perception or production.

CONCLUSIONS

The results suggest a limited and inconsistent transfer of perceptual learning to lexical tone production in children with CIs, which challenges the notion of a robust transfer and highlights the complexity of the interaction between perceptual training and production outcomes. Further research on individual differences with a longitudinal design is needed to optimize the training protocol or tailor interventions to better meet the diverse needs of learners.

Voice Discrimination in Quiet and in Background Noise by Simulated and Real Cochlear Implant Users

PURPOSE

Cochlear implant (CI) users demonstrate poor voice discrimination (VD) in quiet conditions based on the speaker's fundamental frequency (fo) and formant frequencies (i.e., vocal-tract length [VTL]). Our purpose was to examine the effect of background noise at levels that allow good speech recognition thresholds (SRTs) on VD via acoustic CI simulations and CI hearing.

METHOD

Forty-eight normal-hearing (NH) listeners who listened via noise-excited (n = 20) or sinewave (n = 28) vocoders and 10 prelingually deaf CI users (i.e., whose hearing loss began before language acquisition) participated in the study. First, the signal-to-noise ratio (SNR) that yields 70.7% correct SRT was assessed using an adaptive sentence-in-noise test. Next, the CI simulation listeners performed 12 adaptive VDs: six in quiet conditions, two with each cue (fo, VTL, fo + VTL), and six amid speech-shaped noise. The CI participants performed six VDs: one with each cue, in quiet and amid noise. SNR at VD testing was 5 dB higher than the individual's SRT in noise (SRTn +5 dB).

RESULTS

Results showed the following: (a) Better VD was achieved via the noise-excited than the sinewave vocoder, with the noise-excited vocoder better mimicking CI VD; (b) background noise had a limited negative effect on VD, only for the CI simulation listeners; and (c) there was a significant association between SNR at testing and VTL VD only for the CI simulation listeners.

CONCLUSIONS

For NH listeners who listen to CI simulations, noise that allows good SRT can nevertheless impede VD, probably because VD depends more on bottom-up sensory processing. Conversely, for prelingually deaf CI users, noise that allows good SRT hardly affects VD, suggesting that they rely strongly on bottom-up processing for both VD and speech recognition.

Production of Tone 2 in disyllabic words in Mandarin Chinese speaking children aged 3-5 with a cochlear implant and a contralateral hearing aid

To investigate Mandarin Tone 2 production of disyllabic words of prelingually deafened children with a cochlear implant (CI) and a contralateral hearing aid (HA) and to evaluate the relationship between their demographic variables and tone-production ability. Thirty prelingually Mandarin-speaking preschoolers with CI+HA and 30 age-matched normal-hearing (NH) children participated in the study. Fourteen disyllabic words were recorded from each child. A total of 840 tokens (14 × 60) were then used in tone-perception tests in which four speech therapists participated. The production of T2-related disyllabic words of the bimodal group was significantly worse than that of the NH group, as reflected in the overall accuracy (88.57% ± 16.31% vs 99.29% ± 21.79%, p < 0.05), the accuracy of T1+T2 (93.33% vs 100%), the accuracy of T2+T1 (66.67 ± 37.91% vs 98.33 ± 9.13%), and the accuracy of T2+T4 (78.33 ± 33.95% vs 100%). In addition, the bimodal group showed significantly inferior production accuracy of T2+T1 than T2+T2 and T3+T2, p < 0.05. Both bimodal age and implantation age were significantly negatively correlated with the overall production accuracy, p < 0.05. For the error patterns, bimodal participants experienced more errors when T2 was in the first position of the tone combination, and T2 was most likely to be mispronounced as T1 and T3. Bimodal patients aged 3-5 have T2-related disyllabic lexical tone production defects, and their performances are related to tone combination, implantation age, and bimodal age.

Development of Mandarin Lexical Tone Identification in Noise and Its Relation With Working Memory

PURPOSE

This study aimed to examine the developmental trajectory of Mandarin tone identification in quiet and two noisy conditions: speech-shaped noise (SSN) and multitalker babble noise. In addition, we evaluated the relationship between tonal identification development and working memory capacity.

METHOD

Ninety-three typically developing children aged 5-8 years and 23 young adults completed categorical identification of two tonal continua (Tone 1-4 and Tone 2-3) in quiet, SSN, and babble noise. Their working memory was additionally measured using auditory digit span tests. Correlation analyses between digit span scores and boundary widths were performed.

RESULTS

Six-year-old children have achieved the adultlike ability of categorical identification of Tone 1-4 continuum under both types of noise. Moreover, 6-year-old children could identify Tone 2-3 continuum as well as adults in SSN. Nonetheless, the child participants, even 8-year-olds, performed worse when tokens from Tone 2-3 continuum were masked by babble noise. Greater working memory capacity was associated with better tone identification in noise for preschoolers aged 5-6 years; however, for school-age children aged 7-8 years, such correlation only existed in Tone 2-3 continuum in SSN.

CONCLUSIONS

Lexical tone perception might take a prolonged time to achieve adultlike competence in babble noise relative to SSN. Moreover, a significant interaction between masking type and stimulus difficulty was found, as indicated by Tone 2-3 being more susceptible to interference from babble noise than Tone 1-4. Furthermore, correlations between working memory capacity and tone perception in noise varied with developmental stage, stimulus difficulty, and masking type.

Tone perception development in Mandarin-speaking children with cochlear implants

OBJECTIVE

The purpose of this study was to examine the longer-term effects of cochlear implant (CI) use on tone perception by evaluating improvement in Mandarin tone recognition in children with CIs 2 and 3 years post CI activation, and to explore the effects of implant age, chronological age and duration of CI use on the development of tone perception.

METHODS

Tone perception was assessed in 29 bilateral profound hearing impaired children (mean chronological age = 4.6 years, SD = 0.7 years) with unilateral CIs at 24 and 36 months after CI activation using the tone perception subtest in the Mandarin Early Speech Perception (MESP-T) test.

RESULTS

Children's tone recognition for tone pairs and individual tones improved significantly between 2 and 3 years post CI use, showing an increase in average tone recognition score from 73.2% to 81.8%, which was significantly higher than chance level (i.e. 50%). There was no significant correlation between tone recognition ability and either implant age or chronological age at two evaluation points. Further analysis revealed that the tone recognition score for tone pair 2-3 was significantly lower than that for other tone pairs except for tone pair 2-4.

CONCLUSIONS

Longer CI experience can significantly improve tone recognition ability in CI children between 2 and 3 years post CI activation.

Cochlear-implant Mandarin tone recognition with a disyllabic word corpus

Despite pitch being considered the primary cue for discriminating lexical tones, there are secondary cues such as loudness contour and duration, which may allow some cochlear implant (CI) tone discrimination even with severely degraded pitch cues. To isolate pitch cues from other cues, we developed a new disyllabic word stimulus set (Di) whose primary (pitch) and secondary (loudness) cue varied independently. This Di set consists of 270 disyllabic words, each having a distinct meaning depending on the perceived tone. Thus, listeners who hear the primary pitch cue clearly may hear a different meaning from listeners who struggle with the pitch cue and must rely on the secondary loudness contour. A lexical tone recognition experiment was conducted, which compared Di with a monosyllabic set of natural recordings. Seventeen CI users and eight normal-hearing (NH) listeners took part in the experiment. Results showed that CI users had poorer pitch cues encoding and their tone recognition performance was significantly influenced by the “missing” or “confusing” secondary cues with the Di corpus. The pitch-contour-based tone recognition is still far from satisfactory for CI users compared to NH listeners, even if some appear to integrate multiple cues to achieve high scores. This disyllabic corpus could be used to examine the performance of pitch recognition of CI users and the effectiveness of pitch cue enhancement based Mandarin tone enhancement strategies. The Di corpus is freely available online: https://github.com/BetterCI/DiTone.

A Review of Speech Perception of Mandarin-Speaking Children With Cochlear Implantation

Objective: This paper reviewed the literature on the development of and factors affecting speech perception of Mandarin-speaking children with cochlear implantation (CI). We also summarized speech outcome measures in standard Mandarin for evaluating auditory and speech perception of children with CI.

Method: A comprehensive search of Google Scholar and PubMed was conducted from March to June 2021. Search terms used were speech perception/lexical tone recognition/auditory perception AND cochlear implant AND Mandarin/Chinese.

Conclusion: Unilateral CI recipients demonstrated continuous improvements in auditory and speech perception for several years post-activation. Younger age at implantation and longer duration of CI use contribute to better speech perception. Having undergone a hearing aid trial before implantation and having caregivers whose educational level is higher may lead to better performance. While the findings that support the use of CI to improve speech perception continue to grow, much research is needed to validate the use of unilateral and bilateral implantation. Evidence to date, however, revealed bimodal benefits over CI-only conditions in lexical tone recognition and sentence perception in noise. Due to scarcity of research, conclusions on the benefits of bilateral CIs compared to unilateral CI or bimodal CI use cannot be drawn. Therefore, future research on bimodal and bilateral CIs is needed to guide evidence-based clinical practice.

Pitch Accuracy of Vocal Singing in Deaf Children With Bimodal Hearing and Bilateral Cochlear Implants

OBJECTIVES

The purpose of the present study was to investigate the pitch accuracy of vocal singing in children with severe to profound hearing loss who use bilateral cochlear implants (CIs) or bimodal devices [CI at one ear and hearing aid (HA) at the other] in comparison to similarly-aged children with normal-hearing (NH).

DESIGN

The participants included four groups: (1) 26 children with NH, (2) 13 children with bimodal devices, (3) 31 children with bilateral CIs that were implanted sequentially, and (4) 10 children with bilateral CIs that were implanted simultaneously. All participants were aged between 7 and 11 years old. Each participant was recorded singing a self-chosen song that was familiar to him or her. The fundamental frequencies (F0) of individual sung notes were extracted and normalized to facilitate cross-subject comparisons. Pitch accuracy was quantified using four pitch-based metrics calculated with reference to the target music notes: mean note deviation, contour direction, mean interval deviation, and F0 variance ratio. A one-way ANOVA was used to compare listener-group difference on each pitch metric. A principal component analysis showed that the mean note deviation best accounted for pitch accuracy in vocal singing. A regression analysis examined potential predictors of CI children's singing proficiency using mean note deviation as the dependent variable and demographic and audiological factors as independent variables.

RESULTS

The results revealed significantly poorer performance on all four pitch-based metrics in the three groups of children with CIs in comparison to children with NH. No significant differences were found among the three CI groups. Among the children with CIs, variability in the vocal singing proficiency was large. Within the group of 13 bimodal users, the mean note deviation was significantly correlated with their unaided pure-tone average thresholds (r = 0.582, p = 0.037). The regression analysis for all children with CIs, however, revealed no significant demographic or audiological predictor for their vocal singing performance.

CONCLUSION

Vocal singing performance in children with bilateral CIs or bimodal devices is not significantly different from each other on a group level. Compared to children with NH, the pediatric bimodal and bilateral CI users, in general, demonstrated significant deficits in vocal singing ability. Demographic and audiological factors, known from previous studies to be associated with good speech and language development in prelingually-deafened children with CIs, were not associated with singing accuracy for these children.

Dialect Effects on Mandarin Tone Perception Development

This study aimed to characterize factors that influence early dialect development in a language environment with multiple dialects. Children were evaluated for these dialect effects compared with normal hearing referenced measures of speech and language development that are commonly implemented in hearing-impaired children. Dialect exposure and use were assessed longitudinally in Chinese children (2-6 years old) that were raised in a community where Putonghua (PTH) and Sichuanhua (SCH) Mandarin dialects were used. Lexical tones in these dialects are different. A total of 20 boys and 20 girls (2 years old at the beginning of the study) that attended the same nursery school were included in this study. SCH was used by the majority of subjects <4 years old. The majority of subjects >4 years old used either dialect, with a few users of both dialects at this age. PTH tone perception did not differ significantly as a function of dialect use. Tone recognition and discrimination were >90% accurate by 6 years old, in contrast to previous results for children with minimal exposure and use of PTH. Children with approximately ⩾50% PTH exposure might be accurately assessed with norm-referenced speech materials spoken in PTH, regardless of their preferred dialect. However, the current norm-referenced assessments of children with minimal PTH exposure and nonusers of the dialect might be inaccurate.

Voice Emotion Recognition by Mandarin-Speaking Children with Cochlear Implants

Objectives

Emotional expressions are very important in social interactions. Children with cochlear implants can have voice emotion recognition deficits due to device limitations. Mandarin-speaking children with cochlear implants may face greater challenges than those speaking nontonal languages; the pitch information is not well preserved in cochlear implants, and such children could benefit from child-directed speech, which carries more exaggerated distinctive acoustic cues for different emotions. This study investigated voice emotion recognition, using both adult-directed and child-directed materials, in Mandarin-speaking children with cochlear implants compared with normal hearing peers. The authors hypothesized that both the children with cochlear implants and those with normal hearing would perform better with child-directed materials than with adult-directed materials.

Design

Thirty children (7.17-17 years of age) with cochlear implants and 27 children with normal hearing (6.92-17.08 years of age) were recruited in this study. Participants completed a nonverbal reasoning test, speech recognition tests, and a voice emotion recognition task. Children with cochlear implants over the age of 10 years also completed the Chinese version of the Nijmegen Cochlear Implant Questionnaire to evaluate the health-related quality of life. The voice emotion recognition task was a five-alternative, forced-choice paradigm, which contains sentences spoken with five emotions (happy, angry, sad, scared, and neutral) in a child-directed or adult-directed manner.

Results

Acoustic analyses showed substantial variations across emotions in all materials, mainly on measures of mean fundamental frequency and fundamental frequency range. Mandarin-speaking children with cochlear implants displayed a significantly poorer performance than normal hearing peers in voice emotion perception tasks, regardless of whether the performance is measured in accuracy scores, Hu value, or reaction time. Children with cochlear implants and children with normal hearing were mainly affected by the mean fundamental frequency in speech emotion recognition tasks. Chronological age had a significant effect on speech emotion recognition in children with normal hearing; however, there was no significant correlation between chronological age and accuracy scores in speech emotion recognition in children with implants. Significant effects of specific emotion and test materials (better performance with child-directed materials) in both groups of children were observed. Among the children with cochlear implants, age at implantation, percentage scores of nonverbal intelligence quotient test, and sentence recognition threshold in quiet could predict recognition performance in both accuracy scores and Hu values. Time wearing cochlear implant could predict reaction time in emotion perception tasks among children with cochlear implants. No correlation was observed between the accuracy score in voice emotion perception and the self-reported scores of health-related quality of life; however, the latter were significantly correlated with speech recognition skills among Mandarin-speaking children with cochlear implants.

Conclusions

Mandarin-speaking children with cochlear implants could have significant deficits in voice emotion recognition tasks compared with their normally hearing peers and can benefit from the exaggerated prosody of child-directed speech. The effects of age at cochlear implantation, speech and language development, and cognition could play an important role in voice emotion perception by Mandarin-speaking children with cochlear implants.

Deficient sensory and cognitive processing in children with cochlear implants: An event-related potential study

Compared with children having normal hearing (NH), those with cochlear implants (CIs) perform poorly in spoken language comprehension which involves both low-level acoustic encoding and higher-level cognitive processing. Here, we performed an electroencephalography study to portray this brain dynamics of speech perception in CI children. We presented a Mandarin Chinese monosyllable or four-syllable idiom to CI and NH children, and infrequently varied its lexical tone to form a novel monosyllable or pseudo-idiom in an oddball paradigm. The tone contrast embedded in the monosyllables evoked similar mismatch negativities (MMNs) in CI and NH children at an early stage (~200 ms). However, the amplitude of the MMN evoked by the tone contrast in the idiom context was significantly lower in CI children than in NH children. Furthermore, robust late discriminative negativity (LDN) at a late stage (~500 ms) was found only in NH children, but not in CI children. The MMN and LDN findings indicate deficits of low-level acoustic encoding in a complex context (such as an idiom) and higher-level cognitive processing in CI children, respectively. Both deficient sensory and cognitive processing may contribute to the speech perception difficulties in CI children.

Improving the Intelligibility of Speech for Simulated Electric and Acoustic Stimulation Using Fully Convolutional Neural Networks

Combined electric and acoustic stimulation (EAS) has demonstrated better speech recognition than conventional cochlear implant (CI) and yielded satisfactory performance under quiet conditions. However, when noise signals are involved, both the electric signal and the acoustic signal may be distorted, thereby resulting in poor recognition performance. To suppress noise effects, speech enhancement (SE) is a necessary unit in EAS devices. Recently, a time-domain speech enhancement algorithm based on the fully convolutional neural networks (FCN) with a short-time objective intelligibility (STOI)-based objective function (termed FCN(S) in short) has received increasing attention due to its simple structure and effectiveness of restoring clean speech signals from noisy counterparts. With evidence showing the benefits of FCN(S) for normal speech, this study sets out to assess its ability to improve the intelligibility of EAS simulated speech. Objective evaluations and listening tests were conducted to examine the performance of FCN(S) in improving the speech intelligibility of normal and vocoded speech in noisy environments. The experimental results show that, compared with the traditional minimum-mean square-error SE method and the deep denoising autoencoder SE method, FCN(S) can obtain better gain in the speech intelligibility for normal as well as vocoded speech. This study, being the first to evaluate deep learning SE approaches for EAS, confirms that FCN(S) is an effective SE approach that may potentially be integrated into an EAS processor to benefit users in noisy environments.

Acoustic Assessment of Tone Production of Prelingually-Deafened Mandarin-Speaking Children With Cochlear Implants

Objective

The purpose of the present study was to investigate Mandarin tone production performance of prelingually deafened children with cochlear implants (CIs) using modified acoustic analyses and to evaluate the relationship between demographic factors of those CI children and their tone production ability.

Methods

Two hundred seventy-eight prelingually deafened children with CIs and 173 age-matched normal-hearing (NH) children participated in the study. Thirty-six monosyllabic Mandarin Chinese words were recorded from each subject. The fundamental frequencies (F0) were extracted from the tone tokens. Two acoustic measures (i.e., differentiability and hit rate) were computed based on the F0 onset and offset values (i.e., the tone ellipses of the two-dimensional [2D] method) or the F0 onset, midpoint, and offset values (i.e., the tone ellipsoids of the 3D method). The correlations between the acoustic measures as well as between the methods were performed. The relationship between demographic factors and acoustic measures were also explored.

Results

The children with CIs showed significantly poorer performance in tone differentiability and hit rate than the NH children. For both CI and NH groups, performance on the two acoustic measures was highly correlated with each other (r values: 0.895–0.961). The performance between the two methods (i.e., 2D and 3D methods) was also highly correlated (r values: 0.774–0.914). Age at implantation and duration of CI use showed a weak correlation with the scores of acoustic measures under both methods. These two factors jointly accounted for 15.4–18.9% of the total variance of tone production performance.

Conclusion

There were significant deficits in tone production ability in most prelingually deafened children with CIs, even after prolonged use of the devices. The strong correlation between the two methods suggested that the simpler, 2D method seemed to be efficient in acoustic assessment for lexical tones in hearing-impaired children. Age at implantation and especially the duration of CI use were significant, although weak, predictors for tone development in pediatric CI users. Although a large part of tone production ability could not be attributed to these two factors, the results still encourage early implantation and continual CI use for better lexical tone development in Mandarin-speaking pediatric CI users.

A Cross-Language Comparison of Sentence Recognition Using American English and Mandarin Chinese HINT and AzBio Sentences

OBJECTIVES

The aim of this study was to perform a cross-language comparison of two commonly used sentence-recognition materials (i.e., Hearing in Noise Test [HINT] and AzBio) in American English (AE) and Mandarin Chinese (MC).

DESIGNS

Sixty normal-hearing, native English-speaking and 60 normal-hearing, native Chinese-speaking young adults were recruited to participate in three experiments. In each experiment, the subjects were tested in their native language. In experiments I and II, noise and tone vocoders were used to process the HINT and AzBio sentences, respectively. The number of channels varied from 1 to 9, with an envelope cutoff frequency of 160 Hz. In experiment III, the AE AzBio and the MC HINT sentences were tested in speech-shaped noise at various signal to noise ratios (i.e., -20, -15, -10, -5, and 0 dB). The performance-intensity functions of sentence recognition using the two sets of sentence materials were compared.

RESULTS

Results of experiments I and II using vocoder processing indicated that the AE and MC versions of HINT and AzBio sentences differed in level of difficulty. The AE version yielded higher recognition performance than the MC version for both HINT and AzBio sentences. The type of vocoder processing (i.e., tone and noise vocoders) produced little differences in sentence-recognition performance in both languages. Incidentally, the AE AzBio sentences and the MC HINT sentences had similar recognition performance under vocoder processing. Such similarity was further confirmed under noise conditions in experiment III, where the performance-intensity functions of the two sets of sentences were closely matched.

CONCLUSIONS

The HINT and AzBio sentence materials developed in AE and MC differ in level of difficulty. The AE AzBio and the MC HINT sentence materials are similar in level of difficulty. In cross-language comparative research, the MC HINT and the AE AzBio sentences should be chosen for the respective language as the target sentence-recognition test materials.

Factors Affecting Bimodal Benefit in Pediatric Mandarin-Speaking Chinese Cochlear Implant Users

OBJECTIVES

While fundamental frequency (F0) cues are important to both lexical tone perception and multitalker segregation, F0 cues are poorly perceived by cochlear implant (CI) users. Adding low-frequency acoustic hearing via a hearing aid in the contralateral ear may improve CI users' F0 perception. For English-speaking CI users, contralateral acoustic hearing has been shown to improve perception of target speech in noise and in competing talkers. For tonal languages such as Mandarin Chinese, F0 information is lexically meaningful. Given competing F0 information from multiple talkers and lexical tones, contralateral acoustic hearing may be especially beneficial for Mandarin-speaking CI users' perception of competing speech.

DESIGN

Bimodal benefit (CI+hearing aid - CI-only) was evaluated in 11 pediatric Mandarin-speaking Chinese CI users. In experiment 1, speech recognition thresholds (SRTs) were adaptively measured using a modified coordinated response measure test; subjects were required to correctly identify 2 keywords from among 10 choices in each category. SRTs were measured with CI-only or bimodal listening in the presence of steady state noise (SSN) or competing speech with the same (M+M) or different voice gender (M+F). Unaided thresholds in the non-CI ear and demographic factors were compared with speech performance. In experiment 2, SRTs were adaptively measured in SSN for recognition of 5 keywords, a more difficult listening task than the 2-keyword recognition task in experiment 1.

RESULTS

In experiment 1, SRTs were significantly lower for SSN than for competing speech in both the CI-only and bimodal listening conditions. There was no significant difference between CI-only and bimodal listening for SSN and M+F (p > 0.05); SRTs were significantly lower for CI-only than for bimodal listening for M+M (p < 0.05), suggesting bimodal interference. Subjects were able to make use of voice gender differences for bimodal listening (p < 0.05) but not for CI-only listening (p > 0.05). Unaided thresholds in the non-CI ear were positively correlated with bimodal SRTs for M+M (p < 0.006) but not for SSN or M+F. No significant correlations were observed between any demographic variables and SRTs (p > 0.05 in all cases). In experiment 2, SRTs were significantly lower with two than with five keywords (p < 0.05). A significant bimodal benefit was observed only for the 5-keyword condition (p < 0.05).

CONCLUSIONS

With the CI alone, subjects experienced greater interference with competing speech than with SSN and were unable to use voice gender difference to segregate talkers. For the coordinated response measure task, subjects experienced no bimodal benefit and even bimodal interference when competing talkers were the same voice gender. A bimodal benefit in SSN was observed for the five-keyword condition but not for the two-keyword condition, suggesting that bimodal listening may be more beneficial as the difficulty of the listening task increased. The present data suggest that bimodal benefit may depend on the type of masker and/or the difficulty of the listening task.

Mandarin lexical tone recognition in bimodal cochlear implant users

Objective: To assess the recognition of lexical tones in Mandarin-speaking bimodal cochlear implant (CI) subjects.Design: Lexical tone recognition in quiet and noise (SNR= +5 dB) was measured with electric stimulation (CI alone) or bimodal stimulation (CI + hearing aid (HA)). The recognition and confusion rates of the four tones (T1, T2, T3 and T4) were analysed. Spearman correlation analysis was performed to examine the relationship between hearing levels in the contralateral ear and bimodal benefits.Study sample: Twenty native Mandarin-speaking bimodal CI users, with ages ranging from 16-49 years.Results: Relative to the CI alone, mean tone recognition with the CI + HA improved significantly from 84.1-92.1% correct in quiet (+8 points) and from 57.9-73.1% correct in noise (+15.2 points). Tone confusions between T2 and T3 were the most prominent in all test conditions, and T4 tended to be labelled as T3 in noise. There was no significant correlation between the bimodal benefits for tone recognition and the unaided or HA-aided pure-tone thresholds at 0.25 kHz.Conclusion: Listeners with CI + HA exhibited significantly better tone recognition than with CI alone. The bimodal advantage for tone recognition was greater in noise than in quiet, perhaps due to a ceiling effect in quiet.

Health-Related Quality of Life in Mandarin-Speaking Children With Cochlear Implants

OBJECTIVES

The primary aim of this study was to evaluate the health-related quality of life (HRQoL) of children with cochlear implants (CIs) from the parental perspective. The secondary objective was to explore possible relationships between demographic variables (such as age at assessment, gender, age at implantation, and duration of language rehabilitation) and the HRQoL. The third objective was to determine the developmental trajectories of HRQoL.

DESIGN

This study included parents of 123 children with CIs (mean age, 40.45 months; mean age of CI implantation, 24.74 months; mean device experience, 16.34 months). The time periods for follow-up were at 0, 1, 2, 3, 6, and 12-month intervals of CI use. The Mandarin Children with Cochlear Implants: Parental Perspectives questionnaire was employed to assess HRQoL.

RESULTS

Parents were satisfied with HRQoL, especially with the domain of social relations; however, education received a less positive rating. The duration of CI use was positively correlated with 5 domains, suggesting that children who used CIs for a longer time had higher HRQoL ratings. Children with longer language rehabilitation received more positive ratings in the domains of social relations and education (p < 0.05); children whose mothers had higher education levels received more positive ratings in the domain of general functioning (p < 0.05); children living in cities received more positive ratings in the domains of communication, general functioning and self-reliance (p < 0.05). Girls received more positive rating than boys in the domain of well-being (p < 0.05). No significant correlation was found between age at implantation, age at assessment, only child status, and HRQoL. All domains showed clear increases in the duration of CI use; the majority of the domains showed steeper progress over the first 3 months of CI use. Communication exhibited the most rapid progress, with education progressing at a slower rate.

CONCLUSIONS

Parents were satisfied with all domains of HRQoL. Almost all domains exhibited rapid progress over the first 3 months of CI use, with education progressing at a slower rate. This research underscores the importance of language rehabilitation by revealing that strengthening language rehabilitation could be an effective means of improving the HRQoL of children with CIs.

Singing Proficiency of Members of a Choir Formed by Prelingually Deafened Children With Cochlear Implants

Purpose A group of 10 prelingually deafened children with cochlear implants (CIs) formed a choir and received 21 months of formal music training. The purpose of this study was to evaluate the singing proficiency of these children. Method The participants included all choir members (7 girls and 3 boys, mean age of 9.5 years old) who were unilateral CI users. Meanwhile, 8 age-matched children with normal hearing were recruited as controls and were trained on 1 song for 2 weeks. Individual singing samples without instrument accompaniment were recorded from all participants. The singing samples were subject to acoustic analysis in which the fundamental frequency (F0) of each note was extracted and the duration was measured. Five metrics were developed and computed to quantify the accuracy of their pitch and rhythm performance. The 5 metrics included (a) percent correct of F0 contour direction of adjacent notes, (b) mean deviation of the normalized F0 across the notes, (c) mean deviation of the pitch intervals, (d) mean deviation of adjacent note duration ratio, and (e) mean absolute deviation of note duration. Results The choir members with CIs demonstrated high accuracy in both pitch and tempo measures and performed on par with the children with normal hearing. Early start of music training after implantation and use of bimodal hearing contributed to the development of better music ability in these children with CIs. Conclusion These findings indicated that rigorous music training could facilitate high singing proficiency in prelingually deafened children with CIs.

Simulating speech processing with cochlear implants: How does channel interaction affect learning in neural networks?

We introduce a novel machine learning approach for investigating speech processing with cochlear implants (CIs)—prostheses used to replace a damaged inner ear. Concretely, we use a simple perceptron and a deep convolutional network to classify speech spectrograms that are modified to approximate CI-delivered speech. Implant-delivered signals suffer from reduced spectral resolution, chiefly due to a small number of frequency channels and a phenomenon called channel interaction. The latter involves the spread of information from neighboring channels to similar populations of neurons and can be modeled by linearly combining adjacent channels. We find that early during training, this input modification degrades performance if the networks are first pre-trained on high-resolution speech—with a larger number of channels, and without added channel interaction. This suggests that the spectral degradation caused by channel interaction alters the signal to conflict with perceptual expectations acquired from high-resolution speech. We thus predict that a reduction of channel interaction will accelerate learning in CI users who are implanted after having adapted to high-resolution speech during normal hearing. (The code for replicating our experiments is available online: https://github.com/clips/SimulatingCochlearImplants).

The relationships among verbal ability, executive function, and theory of mind in young children with cochlear implants

This study aims to examine the complex relationships among verbal ability (VA), executive function (EF), and theory of mind (ToM) in young Chinese children with cochlear implants (CCI). All participants were tested using a set of nine measures: one VA, one non-VA, three EF, and four ToM. Our study cohort comprised 82 children aged from 3.8 to 6.9 years, including 36 CCI and 46 children with normal hearing (CNH). CNH outperformed CCI on measures of VA, EF, and ToM. One of the EF tasks, inhibitory control, was significantly associated with ToM after controlling for VA. VA was the primary predictor of EF, while inhibitory control significantly predicted ToM. Our findings suggest that inhibitory control explains the association between EF and ToM, thereby supporting the hypothesis that EF may be a prerequisite for ToM.

Spoken word recognition in noise in Mandarin-speaking pediatric cochlear implant users

OBJECTIVE

The purpose of the present study was to compare spoken word recognition performance in the presence of speech spectrum-shaped noise and four-talker babbles in Mandarin-speaking children with cochlear implants (CIs).

METHODS

Participants included 33 children with unilateral CIs (with a mean age of 10.4 ± 2.9 years old and a mean length of CI use of 7.5 ± 3.0 years). The Standard Chinese version of Lexical Neighborhood Test was implemented in quiet, speech-spectrum-shaped noise (SSN), and four-talker babble (FTB). The signal-to-noise ratios (SNRs) were set at +5 and + 10 dB for both types of maskers. Participants responded by verbally repeating each word they heard and the response was scored as the percentage accuracy of recognition performance. A Generalized Linear Model (GLM) fitting, correlational tests, and a two-way repeated-measures ANOVA were conducted on the percent-correct data.

RESULTS

Word recognition in quiet was on average 74.5% correct but dropped to 57.3% and 48.8% correct for SSN and FTB at 10 dB SNR, respectively, and 44.4% and 32.6% correct for SSN and FTB at 5 dB SNR, respectively. In both quiet and noise conditions, the participants showed lower recognition accuracy for the hard words than for the easy words. Disyllabic words were recognized with higher accuracy rates than were the monosyllabic words. The GLM analysis revealed that all four tested factors (masker type, SNR, lexical neighborhood feature, and lexical type) showed significant impacts on word recognition in children with CIs. Word recognition scores in the two types of maskers were significantly correlated for the disyllabic words at both SNRs and monosyllabic words at 10 dB SNR.

CONCLUSIONS

The present study demonstrated that the lexical features such as the lexical neighborhood characteristics and lexical type had significant effects on speech recognition performance in both quiet and noise conditions in pediatric CI users. Children with years of experience of CI use still encountered remarkable difficulties in everyday listening environment although their speech recognition in quiet reached relatively desired level. Fluctuating noise, such as speech babbles, caused greater challenge than steady-state noise for speech recognition in children with CIs.

The effectiveness of sound-processing strategies on tonal language cochlear implant users: A systematic review

IMPORTANCE

Contemporary cochlear implants (CIs) are well established as a technology for people with severe-to-profound sensorineural hearing loss, with their effectiveness having been widely reported. However, for tonal language CI recipients, speech perception remains a challenge: Conventional signal processing strategies have been demonstrated to possibly provide insufficient information to encode tonal cues, and CI recipients have exhibited considerable deficits in tone perception. Thus, some tonal language-oriented sound-processing strategies have been introduced. The effects of available tonal language-oriented strategies on tone perception are reviewed and evaluated in this study. The results may aid in designing and improving tonal language-appropriate sound-processing strategies for CI recipients.

OBJECTIVE

The objective of this systematic review was to investigate the effects of tonal-language-oriented signal processing strategies on tone perception, music perception, word and sentence recognition.

METHODS

To evaluate the effects of tonal language-oriented strategies on tone perception, we conducted a systematic review. We searched for relevant reports dated from January 1979 to July 2017 using PubMed, Cochrane Library, EBSCO, Web of Science, EMBASE, and 4 Chinese periodical databases (CBMdisc, CNKI, VIP, and Wanfang Data).

RESULTS

According to our search strategy, 672 potentially eligible studies were retrieved from the databases, with 12 of these studies included in the final review after a 4-stage selection process. The majority of sound-processing strategies designed for tonal language were HiResolution® with Fidelity 120 (HiRes 120), fine structure processing, temporal fine structure (TFS), and C-tone. Generally, acute or short-term comparisons between the tonal language-oriented strategies and the conventional strategy did not reveal statistically significant differences in speech perception (or show a small improvement). However, a tendency toward improved tone perception and subjectively reported overall preferred sound quality was observed with the tonal language-oriented strategies.

INTERPRETATION

Conventional signal processing strategies typically provided very limited F0 information via temporal envelopes delivered to the stimulating electrodes. In contrast, tonal language-oriented coding strategies attempted to present more spectral information and TFS cues required for tone perception. Thus, a tendency of improved performance in tonal language perception in CI users was shown.

Tone perception in Mandarin-speaking school age children with otitis media with effusion

OBJECTIVES

The present study explored tone perception ability in school age Mandarin-speaking children with otitis media with effusion (OME) in noisy listening environments. The study investigated the interaction effects of noise, tone type, age, and hearing status on monaural tone perception, and assessed the application of a hierarchical clustering algorithm for profiling hearing impairment in children with OME.

METHODS

Forty-one children with normal hearing and normal middle ear status and 84 children with OME with or without hearing loss participated in this study. The children with OME were further divided into two subgroups based on their severity and pattern of hearing loss using a hierarchical clustering algorithm. Monaural tone recognition was measured using a picture-identification test format incorporating six sets of monosyllabic words conveying four lexical tones under speech spectrum noise, with the signal-to-noise ratio (SNR) conditions ranging from -9 to -21 dB.

RESULTS

Linear correlation indicated tone recognition thresholds of children with OME were significantly correlated with age and pure tone hearing thresholds at every frequency tested. Children with hearing thresholds less affected by OME performed similarly to their peers with normal hearing. Tone recognition thresholds of children with auditory status more affected by OME were significantly inferior to those of children with normal hearing or with minor hearing loss. Younger children demonstrated poorer tone recognition performance than older children with OME. A mixed design repeated-measure ANCOVA showed significant main effects of listening condition, hearing status, and tone type on tone recognition. Contrast comparisons revealed that tone recognition scores were significantly better under -12 dB SNR than under -15 dB SNR conditions and tone recognition scores were significantly worse under -18 dB SNR than those obtained under -15 dB SNR conditions. Tone 1 was the easiest tone to identify and Tone 3 was the most difficult tone to identify for all participants, when considering -12, -15, and -18 dB SNR as within-subject variables. The interaction effect between hearing status and tone type indicated that children with greater levels of OME-related hearing loss had more impaired tone perception of Tone 1 and Tone 2 compared to their peers with lesser levels of OME-related hearing loss. However, tone perception of Tone 3 and Tone 4 remained similar among all three groups. Tone 2 and Tone 3 were the most perceptually difficult tones for children with or without OME-related hearing loss in all listening conditions.

CONCLUSIONS

The hierarchical clustering algorithm demonstrated usefulness in risk stratification for tone perception deficiency in children with OME-related hearing loss. There was marked impairment in tone perception in noise for children with greater levels of OME-related hearing loss. Monaural lexical tone perception in younger children was more vulnerable to noise and OME-related hearing loss than that in older children.

Relative contributions of acoustic temporal fine structure and envelope cues for lexical tone perception in noise

Previous studies have shown that lexical tone perception in quiet relies on the acoustic temporal fine structure (TFS) but not on the envelope (E) cues. The contributions of TFS to speech recognition in noise are under debate. In the present study, Mandarin tone tokens were mixed with speech-shaped noise (SSN) or two-talker babble (TTB) at five signal-to-noise ratios (SNRs; -18 to +6 dB). The TFS and E were then extracted from each of the 30 bands using Hilbert transform. Twenty-five combinations of TFS and E from the sound mixtures of the same tone tokens at various SNRs were created. Twenty normal-hearing, native-Mandarin-speaking listeners participated in the tone-recognition test. Results showed that tone-recognition performance improved as the SNRs in either TFS or E increased. The masking effects on tone perception for the TTB were weaker than those for the SSN. For both types of masker, the perceptual weights of TFS and E in tone perception in noise was nearly equivalent, with E playing a slightly greater role than TFS. Thus, the relative contributions of TFS and E cues to lexical tone perception in noise or in competing-talker maskers differ from those in quiet and those to speech perception of non-tonal languages.