Conversations in noise: Multi-stream architecture vs. deep neural network approach to hearing aids
Created
Written by Petri Korhonen, MSc; Francis Kuk, PhD; Christopher Slugocki, PhD; Heidi Peeters, MAIn this article, the speech-in-noise performance of the RealTime Conversation Enhancement (RTCE) technology implemented in Signia Integrated Xperience (IX) is compared against DNN-based noise reduction technology in a recently introduced hearing aid from a competitor. In a study, 20 participants with hearing loss were fitted with both types of hearing aids and tested in a lab setup simulating a noisy conversation with three conversation partners. The results showed superior conversation-in-noise performance of Signia IX. The RTCE technology yielded an average signal-to-noise ratio improvement of between 1.3 and 2.8 dB over the DNN-based technology at different levels of speech understanding. Furthermore, the participants indicated that they were willing to spend 85% longer time in the conversation in the noisy environment with Signia IX than with the competitor hearing aid.