De-Speakerizing Accented ASR: Measuring and Mitigating Speaker Entanglement for Fair, Reliable Recognition

Read full text Summarise with VIRAT
Authors: Jiaen Sun
Cited by: 0
References: 14

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This research statement proposes to measure and mitigate speaker entanglement, where accent features inadvertently encode who is speaking in accented automatic speech recognition (ASR). We argue that entanglement inflates scores under lenient split for the same speaker and worsens fairness gaps across accents, and we outline a parameter-efficient mitigation that combines adversarial de-speakerization with safe conditioning. The plan is grounded in established results in accented ASR, domain-adversarial learning, and parameter-efficient fine-tuning; it is feasible with public datasets and a frozen Whisper backbone, and can potentially guide low-resource data collection.

Fields of Study & Subjects

Speech Recognition and Synthesis Ethics and Social Impacts of AI Speech and Audio Processing

Publication details

Journal / source
Proceedings of the AAAI Conference on Artificial Intelligence
Volume
40
Issue
48
Pages
41519–41521
Publication date
14 March 2026
Language
EN