2 papers
eess.AS2026
Parameter-Efficient Continual Learning for Automatic Speech Recognition
Steven Vander Eeckt, Hugo Van hamme
Speech foundation models enable strong general-purpose ASR and are attractive for downstream adaptation. However, their size and the catastrophic forgetting induced by sequential f…
eess.AS2026
Efficient Rehearsal for Continual Learning in ASR via Singular Value Tuning
Steven Vander Eeckt, Hugo Van hamme
Continual Learning (CL) in Automatic Speech Recognition (ASR) suffers from catastrophic forgetting when adapting to new tasks, domains, or speakers. A common strategy to mitigate t…