notesum.ai

Published at November 21

Listening for Expert Identified Linguistic Features: Assessment of Audio Deepfake Discernment among Undergraduate Students

cs.SD

cs.CY

eess.AS

Released Date: November 21, 2024

Authors: Noshaba N. Bhalli¹, Nehal Naqvi¹, Chloe Evered¹, Christine Mallinson¹, Vandana P. Janeja¹

Aff.: ¹University of Maryland, Baltimore County

Arxiv: http://arxiv.org/abs/2411.14586v1

Group		N	All Clips	Real	Fake	Unsurety
All Students	C	32	0.04*	0.1*	0.03	-0.02
All Students	E	99	0.01	0.03	0	-0.02*
Female	C	7	0.04	0.14	0.01	0.04
Female	E	28	0.01	0.08	-0.01	-0.05
Male	C	22	0.05	0.08	0.04	-0.03
Male	E	69	0.01	0.01	0	-0.02
English First Language	C	23	0.06*	0.1	0.05	-0.02
English First Language	E	83	0.01	0.02	0	-0.03*
English Not First Language	C	7	-0.02	0	-0.03	-0.01
English Not First Language	E	12	0.03	0.08	0.01	-0.03
Fluent in Another Language	C	17	0.03	0.1	0.01	-0.01
Fluent in Another Language	E	41	0	0.01	-0.01	-0.02
Not Fluent in Another Language	C	14	0.05	0.07	0.05	-0.03
Not Fluent in Another Language	E	54	0.01	0.04	0	-0.02
Computing Major	C	21	0.03	0.13*	0.01	-0.03
Computing Major	E	75	0.01	0.03	0	-0.02
Non-Computing Major	C	8	0.11*	0.06	0.13	0.01
Non-Computing Major	E	17	0.01	0	0.01	-0.02