Talk: Transdisciplinarity in Audio Deepfake Discernment with Expert-in-the-Loop AI Models

Opportunities for audio-related misinformation, fraud, and social disruption are now a clear and imminent threat to individuals and society.  Deepfake audio of human speech can be generated with easy access to several readily available AI voice generation tools—even with a very small original sample. Advancements in generative AI far exceed the advancements in the detection of the AI-generated content. Similarly, advancements in detecting text and video deepfakes exceed advancements in audio deepfake detection. Particularly hindering the continuous robust detection of audio deepfakes is that current AI-based detection models lack a full understanding of the inherent variability of language, and the complexities and uniqueness of human speech. This talk will share challenges and the promising potential in recent transdisciplinary work that incorporates linguistic knowledge into AI-based approaches to provide pathways for expert-in-the-loop detection, and to move beyond expert agnostic AI-based methods for more robust and comprehensive audio deepfake detection.

Dr. Vandana Janeja is associate dean for research and faculty development in the College of Engineering and Information Technology and Professor in the Information Systems (IS) department at UMBC. She served as chair of the IS department at UMBC (2019-2023). She is the author of the book Data Analytics for Cybersecurity published by Cambridge press. She heads the Multi Data Lab at UMBC, which brings together projects addressing important societal aspects such as climate change, ethics in data science, misinformation detection, developing research ecosystems, and advancing data science pedagogy, through the lens of her research in data science.  She is the recipient of the 2024 University System of Maryland, Board of Regents Faculty Award for excellence in scholarship or research. She is a member of the UMBC ADVANCE Executive Committee focusing on diversity in STEM and is a member of the ADVANCE leadership cohort (2022-2024), and an ELATES at Drexel leadership fellow (2021-2022). She is a UMBC innovation fellow (2020-2022) advancing the ideas of including ethics in data science. She served as an expert at NSF supporting data science activities in the CISE directorate (CNS: 2018-2021) and as AAAS S&TP fellow in the Office of the Assistant Director in the CISE directorate (2017-2018). During the fellowship, she helped with visioning and coordination of cross directorate activities for Harnessing the Data Revolution Big Idea at NSF, Data Science Corps, Open Knowledge Network, and Cloud Access.

Support for this event is provided in part by the NSF under SFS grant DGE-1753681.  The UMBC Cyber Defense Lab meets biweekly Fridays 12-1pm. All meetings are open to the public.