The role of speaker gender in vocal emotion recognition
Marialucia Cuciniello (Università Degli Studi della Campania 'Luigi Vanvitelli')
Terry Amorese (Università Degli Studi della Campania 'Luigi Vanvitelli')
Anna Alterio (Università Degli Studi della Campania 'Luigi Vanvitelli')
Daniele Pepe (Università Degli Studi della Campania 'Luigi Vanvitelli')
Anna Esposito (Università Degli Studi della Campania 'Luigi Vanvitelli')
Odette Scharenborg (TU Delft - Multimedia Computing)
Gennaro Cordasco (University of Salerno)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
This study investigates the impact of stimulus gender on vocal emotion recognition. To that end, 3 groups of listeners were asked to classify joy, neutral state, fear, anger and sadness in Italian meaningful and nonsensical sentences in 2 settings: a controlled laboratory setting to which a group of Dutch listeners was assigned and a naturalistic setting to which 2 groups respectively of Dutch and Italian listeners were assigned. Two levels of background noise were applied to the sentences to increase the level of task difficulty. Results showed an impact of stimulus gender on emotion classification performance, which seemed dependent on the emotion category: anger was better recognized when spoken by male speakers in both experimental settings, while joy was better recognized when spoken by female speakers by Dutch listeners in the laboratory and Italian listeners in the naturalistic setting, and fear was better recognized when spoken by female speakers in the naturalistic setting.