WS-2.5

LEVERAGING CONFIDENCE MODELS FOR IDENTIFYING CHALLENGING DATA SUBGROUPS IN SPEECH MODELS

Alkis Koudounas, Eliana Pastor, Politecnico di Torino, Italy; Vittorio Mazzia, Manuel Giollo, Thomas Gueudre, Elisa Reale, Amazon Alexa AI, Italy; Giuseppe Attanasio, Bocconi University, Italy; Luca Cagliero, Sandro Cumani, Politecnico di Torino, Italy; Luca de Alfaro, University of California, Santa Cruz, United States of America; Elena Baralis, Politecnico di Torino, Italy; Daniele Amberti, Amazon Alexa AI, Italy

Session:
WS-2: Trustworthy Speech Processing (TSP) Lecture

Track:
Satellite Workshops

Location:
Workshop Lectures

Presentation Time:
Fri, 19 Apr, 14:46 - 15:10 (UTC +9)

Presentation
Discussion
Resources
No resources available.
Session WS-2
WS-2.1: LCANETS++: ROBUST AUDIO CLASSIFICATION USING MULTI-LAYER NEURAL NETWORKS WITH LATERAL COMPETITION
Sayanton V. Dibbo, Dartmouth College and Los Alamos National Laboratory, United States of America; Juston S. Moore, Garrett T. Kenyon, Michael A. Teti, Los Alamos National Laboratory, United States of America
WS-2.2: MULTI-MODAL APPROACHES FOR IMPROVING THE ROBUSTNESS OF AUDIO-BASED COVID-19 DETECTION SYSTEMS
Drew Grant, Helena Hahn, Adebayo Eisape, Valerie Rennoll, James West, Johns Hopkins University, United States of America
WS-2.3: IMPROVING MEMBERSHIP INFERENCE IN ASR MODEL AUDITING WITH PERTURBED LOSS FEATURES
Francisco Teixeira, INESC-ID/Técnico, University of Lisbon, Portugal; Karla Pizzi, Technical University Munich / Fraunhofer AISEC, Germany; Raphael Olivier, Vatic Labs, United States of America; Alberto Abad, INESC-ID/Técnico, University of Lisbon, Portugal; Bhiksha Raj, LTI, Carnegie Mellon University, United States of America; Isabel Trancoso, INESC-ID/Técnico, University of Lisbon, Portugal
WS-2.4: Characterizing the temporal dynamics of universal speech representations for generalizable deepfake detection
Yi Zhu, Institut national de la recherche scientifique, Canada; Saurabh Powar, Veermata Jijabai Technological Institute, India; Tiago Falk, Institut national de la recherche scientifique, Canada
WS-2.5: LEVERAGING CONFIDENCE MODELS FOR IDENTIFYING CHALLENGING DATA SUBGROUPS IN SPEECH MODELS
Alkis Koudounas, Eliana Pastor, Politecnico di Torino, Italy; Vittorio Mazzia, Manuel Giollo, Thomas Gueudre, Elisa Reale, Amazon Alexa AI, Italy; Giuseppe Attanasio, Bocconi University, Italy; Luca Cagliero, Sandro Cumani, Politecnico di Torino, Italy; Luca de Alfaro, University of California, Santa Cruz, United States of America; Elena Baralis, Politecnico di Torino, Italy; Daniele Amberti, Amazon Alexa AI, Italy
Contacts