Date & Time: June 13th, 9:00-10:00
Location: Seminar Room M 3170-73 at Dept. of Automatic Control, LTH
Author: Klara Almgren and Annie Mentzer
Title: Neural Speech Tracking in EEG: Integrating Acoustics and Linguistics for Hearing Aid Users
Supervisor: Bo Bernhardsson, LTH,  Martin Skoglund and Emina Alickovic (Eriksholm)
Examiner: Pontus Giselsson

Abstract: 

This master thesis  explores the neural encoding of different speech features in hearing aid users’ brain response using EEG, TRFs and multivariate TRFs. The primary objective is to investigate how different acoustic and linguistic features of speech are represented in the brain in a challenging listening scenario, and how these representations are influenced by noise reduction schemes. The acoustic features (e.g., audio envelope), phonetic features (e.g., phoneme and word onsets), as well as the linguistic features (e.g., word surprisal derived from a large language model (GPT-2)) were analyzed.