Effect of masker type and age on speech intelligibility and spatial release from masking in children and adults

Listening to speech in competing sounds poses a major difficulty for children with impaired hearing. This study aimed to determine the ability of children (3-12 yr of age) to use spatial separation between target speech and competing babble to improve speech intelligibility. Fifty-eight children (31 with normal hearing and 27 with impaired hearing who use bilateral hearing aids) were assessed by word and sentence material. Speech reception thresholds (SRTs) were measured with speech presented from 0 azimuth, and competing babble from either 0 or 690 azimuth. Spatial release from masking (SRM) was defined as the difference between SRTs measured with co-located speech and babble and SRTs measured with spatially separated speech and babble. On average, hearingimpaired children attained near-normal performance when speech and babble originated from the frontal source, but performed poorer than their normal-hearing peers when babble was spatially separated from target speech. On average, normal-hearing children obtained an SRM of 3 dB whereas children with hearing loss did not demonstrate SRM. Results suggest that hearing-impaired children may need enhancement in signal-to-noise ratio to hear speech in difficult listening conditions as well as normal-hearing children.

Section: B Srm: Effect Of Age and Hearing Thresholdssupporting

confidence: 73%

Spatial release from masking in normal-hearing children and children who use hearing aids

Ching

Wanrooy

Dillon

et al. 2011

“…It is documented that children show elevated speech reception thresholds relative to adults when recognizing noisy speech (Hall et al, 2002;Johnstone and Litovsky, 2006). Musicians exhibit higher intelligibility in speech-innoise conditions compared to non-musicians (Parbery-Clark et al, 2009), presumably, because of their more extensive experience in listening to polyphonic signals.…”

Section: Discussionmentioning

confidence: 99%

Large-scale training to increase speech intelligibility for hearing-impaired listeners in novel noises

Chen

Wang

Yoho

et al. 2016

167

127

Supervised speech segregation has been recently shown to improve human speech intelligibility in noise, when trained and tested on similar noises. However, a major challenge involves the ability to generalize to entirely novel noises. Such generalization would enable hearing aid and cochlear implant users to improve speech intelligibility in unknown noisy environments. This challenge is addressed in the current study through large-scale training. Specifically, a deep neural network (DNN) was trained on 10 000 noises to estimate the ideal ratio mask, and then employed to separate sentences from completely new noises (cafeteria and babble) at several signal-to-noise ratios (SNRs). Although the DNN was trained at the fixed SNR of À 2 dB, testing using hearing-impaired listeners demonstrated that speech intelligibility increased substantially following speech segregation using the novel noises and unmatched SNR conditions of 0 dB and 5 dB. Sentence intelligibility benefit was also observed for normal-hearing listeners in most noisy conditions. The results indicate that DNN-based supervised speech segregation with large-scale training is a very promising approach for generalization to new acoustic environments.

“…SRTs were calculated using Maximum Likelihood Estimation (MLE) methods, which define threshold as the point on the psychometric function at which the target intensity corresponded to 79.4% correct. This method is identical to the methods used previously by Litovsky and colleagues (Garadat and Litovsky, 2007;Johnstone and Litovsky, 2006;Litovsky, 2005;Misurelli and Litovsky, 2012). , such that,…”

Section: E Speech Reception Threshold Estimationmentioning

confidence: 99%

“…The experimental test consisted of a 4-alternativeforced-choice task (Garadat and Litovsky, 2007;Johnstone and Litovsky, 2006;Litovsky, 2005;Litovsky et al, 2006;Misurelli and Litovsky, 2012), in which the target spondee was identified from four possible icons. On trials with male interferers, listeners were provided with experience hearing the interferer sentences and the male target; subsequently they were instructed to ignore the "boy talkers" and to pay attention to the man's voice (male target).…”

Section: Design and Proceduresmentioning

confidence: 99%

“…Unlike the plethora of data in NH adults, studies investigating the cocktail party problem in children remain small in number. Also notable is that, to date, these studies have used target-interferer stimuli that consist of noise, or different-sex talkers, under the presumption that clarity of the task for the children (i.e., instructions to "listen to the female and ignore the male") would be maximized (Garadat and Litovsky, 2007;Johnstone and Litovsky, 2006;Litovsky, 2005;Litovsky et al, 2006;Misurelli and Litovsky, 2012;Runge et al, 2011).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Spatial release from masking in children with bilateral cochlear implants and with normal hearing: Effect of target-interferer similarity

Misurelli

Litovsky

2015

Self Cite

In complex auditory environments, it is often difficult to separate a target talker from interfering speech. For normal hearing (NH) adult listeners, similarity between the target and interfering speech leads to increased difficulty in separating them; that is, informational masking occurs due to confusability of the target and interferers. This study investigated performance of children with bilateral cochlear implants (BiCIs) when target and interferers were either same-sex (male) talkers, or different-sex talkers (male target, female interferer). Comparisons between children with BiCIs and NH, when matched for age, were also conducted. Speech intelligibility was measured for target and interferers spatially co-located, or spatially separated with the interferers positioned symmetrically (þ90 and À90 ) or asymmetrically (both at þ90 , right). Spatial release from masking (SRM) was computed as the difference between co-located and separated conditions. Within group BiCI comparisons revealed that in the co-located condition speech intelligibility was worse with the same-sex vs different-sex stimuli. There was also a trend for more SRM with the same-sex vs different-sex stimuli. When comparing BiCI to NH listeners, SRM was larger for the NH groups, suggesting that NH children are better able to make use of spatial cues to improve speech understanding in noise.