Longitudinal Speech Recognition in Noise in Children: Effects of Hearing Status and Vocabulary

Walker, Elizabeth; Sapp, Caitlin; Oleson, Jacob; McCreery, Ryan W.

doi:10.3389/fpsyg.2019.02421

Cited by 33 publications

(32 citation statements)

References 65 publications

(121 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Regarding the effects of age, this study confirmed that younger children struggle more to understand speech in noise compared with older ones. This is consistent with a large body of research comparing the effects of noise on younger versus older children (e.g., Johnson, 2000; Soli & Sullivan, 1997; Walker et al., 2019; Werner & Boike, 2001), and can be traced back to developmental changes in terms of language skills of school-age children (e.g., Firmansyah, 2018). At a more general level, this finding implies that since younger children understand speech especially poorly in noisy environments it is particularly important to provide classrooms with good acoustical conditions for them.…”

Section: Discussionsupporting

confidence: 85%

A Danish Sentence Corpus for Assessing Speech Recognition in Noise in School-Age Children

et al. 2020

View full text Add to dashboard Cite

For the audiological assessment of the speech-in-noise abilities of children with normal or impaired hearing, appropriate test materials are required. However, in Denmark, no standardized materials exist. The purpose of this study was to develop a Danish sentence corpus suitable for testing school-age children. Based on the 600 validated test sentences from the Danish DAT ( Dagmar, Asta, or Tine) corpus, 11 test lists comprising 20 sentences each were carefully constructed. These lists were evaluated in terms of their perceptual similarity and reliability with a group of 20 typically developing, normal-hearing children aged 6 to 12 years. Using stationary speech-shaped noise and diotic stimulus presentation, speech recognition thresholds (SRTs) were measured twice per list and participant at two separate visits. The analyses showed that six test lists were perceptually equivalent. These lists are characterized by a grand average SRT of −2.6 dB signal-to-noise ratio, a test–retest improvement of 0.6 dB, and a within-subject standard deviation of 1.1 dB signal-to-noise ratio. The other lists were characterized by slightly higher SRTs, slightly larger training effects, and slightly larger measurement uncertainty, but were otherwise also usable. Overall, it is therefore concluded that the developed corpus is suited for assessing speech recognition in noise in Danish 6- to 12-year olds. The corpus is publicly available.

show abstract

Section: Discussionsupporting

confidence: 85%

A Danish Sentence Corpus for Assessing Speech Recognition in Noise in School-Age Children

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Tomblin et al (2015) found that children who wore hearing aids for at least 10 hours/day during the preschool years had stronger language growth trajectories than peers who wore hearing aids for less than 10 hours/day. More recently, Walker et al (2019) showed that children with higher amounts of auditory dosage, a measure that combines both a child’s aided audibility for speech sounds and their device use, had better sentence recognition than peers with lower auditory dosage. Consistent auditory access, as measured by hours of device use, predicts individual differences in language and speech perception outcomes for CHL.…”

Section: Introductionmentioning

confidence: 99%

Amount of Hearing Aid Use Impacts Neural Oscillatory Dynamics Underlying Verbal Working Memory Processing for Children With Hearing Loss

et al. 2021

Self Cite

View full text Add to dashboard Cite

Objectives: Children with hearing loss (CHL) may exhibit spoken language delays and may also experience deficits in other cognitive domains including working memory. Consistent hearing aid use (i.e., more than 10 hours per day) ameliorates these language delays; however, the impact of hearing aid intervention on the neural dynamics serving working memory remains unknown. The objective of this study was to examine the association between the amount of hearing aid use and neural oscillatory activity during verbal working memory processing in children with mild-to-severe hearing loss. Design: Twenty-three CHL between 8 and 15 years-old performed a letter-based Sternberg working memory task during magnetoencephalography (MEG). Guardians also completed a questionnaire describing the participants’ daily hearing aid use. Each participant’s MEG data was coregistered to their structural MRI, epoched, and transformed into the time–frequency domain using complex demodulation. Significant oscillatory responses corresponding to working memory encoding and maintenance were independently imaged using beamforming. Finally, these whole-brain source images were correlated with the total number of hours of weekly hearing aid use, controlling for degree of hearing loss. Results: During the encoding period, hearing aid use negatively correlated with alpha-beta oscillatory activity in the bilateral occipital cortices and right precentral gyrus. In the occipital cortices, this relationship suggested that with greater hearing aid use, there was a larger suppression of occipital activity (i.e., more negative relative to baseline). In the precentral gyrus, greater hearing aid use was related to less synchronous activity (i.e., less positive relative to baseline). During the maintenance period, hearing aid use significantly correlated with alpha activity in the right prefrontal cortex, such that with greater hearing aid use, there was less right prefrontal maintenance-related activity (i.e., less positive relative to baseline). Conclusions: This study is the first to investigate the impact of hearing aid use on the neural dynamics that underlie working memory function. These data show robust relationships between the amount of hearing aid use and phase-specific neural patterns during working memory encoding and maintenance after controlling for degree of hearing loss. Furthermore, our data demonstrate that wearing hearing aids for more than ~8.5 hours/day may serve to normalize these neural patterns. This study also demonstrates the potential for neuroimaging to help determine the locus of variability in outcomes in CHL.

show abstract

“…Classrooms are an example of environments where children are subject to noise exposure [3][4][5][6] and where ASR technology is increasingly used in education. The degradation of ASR systems in recognition of noisy children's speech depends on many issues such as noise type, signal-to-noise ratio (SNR), acoustic and linguistic differences in fundamental frequency (F 0 ), speaking rate, and formant frequencies between adult and child speech [7][8][9][10][11][12][13][14][15][16]. Decreased ASR performance of children's speech compared to adults' speech is also explained by the fact that the number of publicly available training data for children's speech (tens of hours) [17,18] is much smaller compared to that of adults' speech (thousands of hours) [19,20].…”

Section: Introductionmentioning

confidence: 99%

Using Data Augmentation and Time-Scale Modification to Improve ASR of Children’s Speech in Noisy Environments

et al. 2021

View full text Add to dashboard Cite

Current ASR systems show poor performance in recognition of children’s speech in noisy environments because recognizers are typically trained with clean adults’ speech and therefore there are two mismatches between training and testing phases (i.e., clean speech in training vs. noisy speech in testing and adult speech in training vs. child speech in testing). This article studies methods to tackle the effects of these two mismatches in recognition of noisy children’s speech by investigating two techniques: data augmentation and time-scale modification. In the former, clean training data of adult speakers are corrupted with additive noise in order to obtain training data that better correspond to the noisy testing conditions. In the latter, the fundamental frequency (F0) and speaking rate of children’s speech are modified in the testing phase in order to reduce differences in the prosodic characteristics between the testing data of child speakers and the training data of adult speakers. A standard ASR system based on DNN–HMM was built and the effects of data augmentation, F0 modification, and speaking rate modification on word error rate (WER) were evaluated first separately and then by combining all three techniques. The experiments were conducted using children’s speech corrupted with additive noise of four different noise types in four different signal-to-noise (SNR) categories. The results show that the combination of all three techniques yielded the best ASR performance. As an example, the WER value averaged over all four noise types in the SNR category of 5 dB dropped from 32.30% to 12.09% when the baseline system, in which no data augmentation or time-scale modification were used, was replaced with a recognizer that was built using a combination of all three techniques. In summary, in recognizing noisy children’s speech with ASR systems trained with clean adult speech, considerable improvements in the recognition performance can be achieved by combining data augmentation based on noise addition in the system training phase and time-scale modification based on modifying F0 and speaking rate of children’s speech in the testing phase.

show abstract

Longitudinal Speech Recognition in Noise in Children: Effects of Hearing Status and Vocabulary

Cited by 33 publications

References 65 publications

A Danish Sentence Corpus for Assessing Speech Recognition in Noise in School-Age Children

A Danish Sentence Corpus for Assessing Speech Recognition in Noise in School-Age Children

Amount of Hearing Aid Use Impacts Neural Oscillatory Dynamics Underlying Verbal Working Memory Processing for Children With Hearing Loss

Using Data Augmentation and Time-Scale Modification to Improve ASR of Children’s Speech in Noisy Environments

Contact Info

Product

Resources

About