Print Email Facebook Twitter Effects of a WPE Dereverberation Algorithm on Speech Perception in Cochlear Implant Users Title Effects of a WPE Dereverberation Algorithm on Speech Perception in Cochlear Implant Users Author van Marrewijk, Esther (TU Delft Applied Sciences; TU Delft Mechanical, Maritime and Materials Engineering) Contributor Stronks, HC (mentor) Frijns, JHM (graduation committee) Langerak, NC (graduation committee) Degree granting institution Delft University of TechnologyUniversiteit LeidenErasmus Universiteit Rotterdam Programme Technical Medicine | Sensing and Stimulation Date 2023-11-22 Abstract IntroductionReverberation causes a reduced speech perception in cochlear implant (CI) users. Therefore, a dereverberation method was proposed based on the weighted prediction error (WPE) algorithm with input from a deep neural network. The speech intelligibility for normal and hearing-aided listeners is improved by cancelling the late reverberations. By addition of a post filtering (PF) part, the WPEPF algorithm aims to delete the early reverberations from the signal, because they have no benefit to CI users. The goal of this study is to measure the effect of this WPE(PF) algorithm on speech intelligibility and sound pleasantness in CI users.MethodsSentences from the Flemish/Dutch Matrix test were spoiled with reverberation. The WPE and WPEPF algorithm were introduced to these sentences. The speech intelligibility on each of these was measured, as well as the sound pleasantness with a MATLAB app. Moreover, the predicted speech intelligibility was calculated using the short-time intelligibility objective.ResultsThe measured speech intelligibility of the reverberated speech with WPE and WPEPF was significantly lower than in reverberation ( p < 0. 0 0 1). The sound pleasantness assessment indicated a lower pleasantness for the algorithms. The predicted speech intelligibility showed that the reverberated speech with algorithm performed worse in the first 0.7 seconds, but outperformed the reverberated speech after that. The word-by-word analysis showed that all words are less intelligible in the reverberated speech with WPE(PF), but especially the first word (38.7 and 38.1\% less).ConclusionWhen used as a dereverberation algorithm, the WPE and WPEPF algorithm with input from a deep neural network had a significantly negative effect on measured speech intelligibility in cochlear implant (CI) users. The algorithms distorted the start of the sentences, as described by the word-to-word analysis and the predicted speech intelligibility calculations. However, the algorithms had no effect on speech intelligibility when applied to clean speech. Subject Cochlear implantDereverberationPre-ProcessingSpeech IntelligibilitySound Pleasantness To reference this document use: http://resolver.tudelft.nl/uuid:35ca5315-3c49-4c65-bf5c-be7780212ea7 Embargo date 2024-11-22 Part of collection Student theses Document type master thesis Rights © 2023 Esther van Marrewijk Files file embargo until 2024-11-22