Fact · Dictation
The authors trace the speaker-group gap to the acoustic models used by the ASR systems.
Checked against its source on October 16, 2026
In the source's own words
“We trace these disparities to the underlying acoustic models used by the ASR systems”
Racial disparities in automated speech recognition, Allison Koenecke et al., Proceedings of the National Academy of Sciences, 2020, Abstract.
Open the original (DOI 10.1073/pnas.1915768117)- What it measures
- The authors found the gap was equally large on identical phrases spoken by Black and white participants.
- What it does not show
- The authors' conclusion from their experiments. It does not say how large the gap is in current systems.
Source
Cited on
- Speech Recognition: What It Is and How Well It WorksTopic
- Dictation Languages, Accents and Voice TranslationTopic
- Google Docs Voice Typing: How to Use It and Its LimitsGuide
- Speech Recognition and Accents: Accuracy and FixesGuide
- Dictation Mistakes and How to Make Voice Typing AccurateGuide
- Is speech recognition less accurate for some people?Question
- Why doesn't voice recognition understand my accent?Question
- Acoustic modelGlossary