Inclusivity of AI Speech in Healthcare: A Decade Look Back

TL;DR

The paper addresses inequities in AI speech technologies for healthcare, focusing on language, demographic diversity, and accessibility for speech impairments. It analyzes dataset inclusivity (via 2015–2024 ASR/TTS datasets) and research trends (via OpenAlex) to reveal gaps and progress. Key findings include a persistent English bias, sparse coverage of African and Indigenous languages, limited metadata on gender and age, and almost no samples from individuals with speech disorders, indicating inclusivity remains a minority focus. The work advocates expansive, ethically collected multilingual datasets and robust policy frameworks to ensure equitable access to AI speech healthcare tools.

Abstract

The integration of AI speech recognition technologies into healthcare has the potential to revolutionize clinical workflows and patient-provider communication. However, this study reveals significant gaps in inclusivity, with datasets and research disproportionately favouring high-resource languages, standardized accents, and narrow demographic groups. These biases risk perpetuating healthcare disparities, as AI systems may misinterpret speech from marginalized groups. This paper highlights the urgent need for inclusive dataset design, bias mitigation research, and policy frameworks to ensure equitable access to AI speech technologies in healthcare.