Language technology practitioners as language managers: arbitrating data bias and predictive bias in ASR
It addresses bias in ASR systems for marginalized language users, offering a language policy perspective for practitioners, but is incremental in its analytical approach.
The paper analyzes how training and testing practices in automatic speech recognition (ASR) lead to data bias, causing worse performance on non-standardized and marginalized language varieties, and proposes reframing language resources as public infrastructure for speech communities.
Despite the fact that variation is a fundamental characteristic of natural language, automatic speech recognition systems perform systematically worse on non-standardised and marginalised language varieties. In this paper we use the lens of language policy to analyse how current practices in training and testing ASR systems in industry lead to the data bias giving rise to these systematic error differences. We believe that this is a useful perspective for speech and language technology practitioners to understand the origins and harms of algorithmic bias, and how they can mitigate it. We also propose a re-framing of language resources as (public) infrastructure which should not solely be designed for markets, but for, and with meaningful cooperation of, speech communities.