Linguistic Distance Segregates Latent Representations in Automatic Speech Recognition Systems
cs.CL, cs.LG
Submitted: 2026-08-31
Updated: 2026-08-31
Comments: to appear in EMNLP finding 2026
Code: https://github.com/tinghui8576/LD-Bias-EngASR
License: http://creativecommons.org/licenses/by/4.0/
The gist: While automatic speech recognition (ASR) models have achieved remarkable improvements in recent years, performance disparities persist across different speaker populations.
Terminology
Abstract
While automatic speech recognition (ASR) models have achieved remarkable improvements in recent years, performance disparities persist across different speaker populations. One such disparity is for speakers whose first languages (L1) are from families distant from English. This paper investigates the relationship between first language background and English ASR performance. Through empirical analysis, we observe that the correlation between speakers' L1 distance and ASR error rates yields a systematic effect on English Speech, with its strength varying across datasets and models. This association is statistically significant in a follow-up analysis accounting for dataset-level variation in Tweedie mixed-effects models (p<0.001 across evaluated models). In addition, analysis of the latent space reveals a L1-based spatial segregation across deeper acoustic layers in the majority of evaluated architectures
Sources
- You don't understand me!: Comparing ASR results for L1 and L2 speakers of Swedish
- Global Performance Disparities Between English-Language Accents in Automatic Speech Recognition
- Evaluating User Perception of Speech Recognition System Quality with Semantic Distance Metric
- Canary-1B-v2 & Parakeet-TDT-0.6B-v3: Efficient and High-Performance Models for Multilingual ASR and AST
- Towards measuring fairness in speech recognition: Fair-Speech dataset
Related papers
- Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
- Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
- Subliminal Steering: Stronger Encoding of Hidden Signals
- MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
- The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
- Untangling the Mechanisms of Misleading Context in Medical Question Answering