Neural correlates of Affective States and Speech Perception: Towards Quality-of- Experience Measurement of Reverberant Speech

Abstract

Traditional speech quality assessments rely on subjective listener evaluations, which are valid but offer limited insight into the perceptual processes involved. Since reverberation significantly affects perceived speech quality and intelligibility—such as during conference calls using loudspeakers—this study investigates the use of electroencephalography (EEG) as an objective tool for detecting degradation due to high reverberation. EEG data, particularly event-related potentials (ERPs), were collected from 22 participants to assess their responses to stimuli with varying reverberation times. Preliminary findings suggest that EEG can detect quality degradation without requiring explicit subjective feedback. Correlations between ERP features and stimulus characteristics indicate that reverberation-induced quality changes are reflected in conscious auditory processing. Overall, ERP analysis proves to be a valid and promising method for objectively evaluating speech quality, especially regarding room acoustics and reverberation effects.