Development of a training module to improve speech perception in noise for adults

Abstract Background and objectives The present study aimed to develop a training module for adults to improve speech-in-noise (SPIN) ability and to derive normative scores for this program in adults. Subjects and methods The study was carried out in two phases: 1) the development of a training module and 2) the administration of the training module to normal-hearing individuals to derive normative cut-off scores. In the initial phase, a material containing lists of 50 activities in the Tamil language to improve SPIN using different signal-to-noise ratios (SNR) from 15 dB SNR to -5 dB SNR, including various types of noise such as environmental noise (fan noise), white noise, single speech babble, multi-speech babble, and cafeteria noise for sentences and words, was developed. These recorded materials were administered to a group of 36 normal-hearing individuals within the age range of 18 to 30 years in the second phase. The mean and the confidence interval for each activity were analyzed to derive cut-off scores. Results The overall mean SPIN scores decreased with the reduction of SNR. The mean SPIN scores were more than 90% for up to +5 dB SNR and reduced to between 80 and 90% at 0 dB SNR; scores further dropped to between 60 and 80% at -5 dB SNR. The participants reported more difficulty with the cafeteria noise, single-speech babble, and multi-speech babble. At the lowest SNRs of 0 dB and -5 dB, participants perceived sentences easier than words. The obtained 90% confidence interval for each activity was used as a cutoff score. The score was used as a reference to move to the next activity. Conclusion Cut-off scores of activities listed in training modules vary depending upon the SNR, type of stimuli, and noise in the normal hearing population. This indicates the necessity for activity-specific normative cut-offs to maintain the motivation level of participants.