analiza aiqbench
This study investigates whether regularization techniques like EWC and mini-KFAC are necessary for adapting ASR models to financial earnings calls, finding that task complementarity often negates the need for such constraints. When forgetting does occur, the research demonstrates that layer-wise mini-KFAC provides more stable optimization and lower word error rates compared to standard diagonal EWC. The findings offer practical guidance for deploying continual learning strategies in specialized domain adaptation without unnecessary regularization overhead.
Does Financial Speech Need Continual Learning Regularization? A Study of EWC and mini-KFAC on Earnings Call ASR
arXiv:2610.13432v1 Announce Type: cross Abstract: Earnings call transcription presents a realistic continual learning scenario where Automatic Speech Recognition (ASR) systems are updated with new recordings but risk catastrophic forgetting of previously learned financial speech patterns. Adapting Whisper-small from a general-domain speech corpus to the Earnings22 benchmark reveals the opposite: all evaluated methods exhibit improvement on held-out financial speech from the same domain during adaptation rather than forgetting. This task complementarity makes regularization counterproductive. Nevertheless, when regularization is warranted, a layer-wise mini-Kronecker-Factored Approximate Curvature (mini-KFAC) approximation seems to achieve lower word error rates (WERs) and more stable optimization than standard diagonal Elastic Weight Consolidation (EWC). The results suggest that task complementarity should be assessed before applying continual learning constraints, and prefer mini-KFAC EWC over diagonal EWC when needed.
Abstract:Earnings call transcription presents a realistic continual learning scenario where Automatic Speech Recognition (ASR) systems are updated with new recordings but risk catastrophic forgetting of previously learned financial speech patterns. Adapting Whisper-small from a general-domain speech corpus to the Earnings22 benchmark reveals the opposite: all evaluated methods exhibit improvement on held-out financial speech from the same domain during adaptation rather than forgetting. This task complementarity makes regularization counterproductive. Nevertheless, when regularization is warranted, a layer-wise mini-Kronecker-Factored Approximate Curvature (mini-KFAC) approximation seems to achieve lower word error rates (WERs) and more stable optimization than standard diagonal Elastic Weight Consolidation (EWC). The results suggest that task complementarity should be assessed before applying continual learning constraints, and prefer mini-KFAC EWC over diagonal EWC when needed.
10 pages, 7 figures, 3 tables. Presented at the 9th International Conference on Natural Language and Speech Processing (ICNLSP 2026), University of Trento, Trento, Italy, September 25-26, 2026. Proceedings to appear in the ACL Anthology
Audio and Speech Processing (eess.AS); Computation and Language (cs.CL); Machine Learning (cs.LG)
arXiv:2610.13432 [eess.AS]
(or arXiv:2610.13432v1 [eess.AS] for this version)
https://doi.org/10.48550/arXiv.2610.13432 arXiv-issued DOI via DataCite (pending registration)
Text integral sub licența deschisă a publicației · 184 cuvinte
Subiecte conexe
BRICS Draws a New Global Line: Full Text of New Delhi Declaration
slguardian.org · 12 sept. 2026
The latest AI news we announced in June 2026
Google · 1 iul. 2026
Elon Musk is steamrolling Wall Street to become a trillionaire
The Verge · 4 iun. 2026
The J-Space Debate, Agent Swarms, and Pacing Frontier AI - Digital Minds Newsletter #4
LessWrong · 18 sept. 2026