Improvements in Pediatric Speech Perception in Noise Using Classifier-Based Noise Management

Purpose: The main objective of this study was to examine the effect of a classifier-based noise management algorithm (AutoSense Sky OS 3.0) on speech perception in noise in pediatric cochlear implant (CI) recipients and to compare the speech perception outcomes with typically hearing peers. Method: This prospective observational study included nine children (ages 9–15 years) with bilateral CIs and nine age-matched, typically hearing peers. Speech perception outcomes were measured in noise using Pediatric AzBio sentences to compare performance in the omnidirectional microphone mode with the classifier-based automatic noise management mode at four signal-to-noise ratios. Ratings of listening ease and speech clarity were recorded to obtain subjective measures of benefit. Results from the bilateral CI recipients were compared to a group of typically hearing children. Paired t tests were used to evaluate outcomes. Results: Classifier-based noise management improved speech perception in noise compared to listening in the omnidirectional microphone mode. CI group averages revealed a 21.4 percentage point and 47.1 percentage point improvement in speech perception when using the Sky OS 3.0 algorithm at 0 and −5 dB SNR, respectively. Listening ease and speech clarity were improved when using the classifier-based noise management algorithm. Conclusions: Pediatric CI recipients can benefit from classifier-based noise management. Given the difficult listening environments children face, especially in typical classroom settings while listening for learning, audiologists should consider activating classifier-based noise management in this population.