Speaking of accent: A content analysis of accent misconceptions in ASR research
Open accessAbstract
Automatic speech recognition (ASR) researchers are working to address the differing transcription performance of ASR by accent or dialect. However, research often has a limited view of accent in ways that reproduce discrimination and limit the scope of potential solutions. In this paper we present a content analysis of 22 papers published in 2022 in top conferences and journals on the topic of accent and ASR. We report on how accent is sometimes mistakenly viewed as something some people don’t have; as having a default; and being an attribute only of the speaker, and not of the listener. We discuss the implications on research and provide recommendations to researchers who hope to reduce ASR biases by accent.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Other research has uncovered the ambiguity and uncertainty inherent in socially constructed labels, such as gender, accent, and content toxicity.
“Other research has uncovered the ambiguity and uncertainty inherent in socially constructed labels, such as gender [85], accent [81], and content toxicity [1, 16].”
From Disability-First AI Dataset Annotation: Co-designing Stuttered Speech Annotation Guidelines with People Who Stutter (2026) · cites this paper as [81]
Cites
Code Saturation Versus Meaning Saturation: How Many Interviews Are Enough?
Hennink, M. M., Kaiser, B. N., & Marconi, V. C. (2017) · Qualitative Health Research · doi:10.1177/1049732316665344 · Cited by 4,597 · see citing works
Gebru, T., Morgenstern, J., Vecchione, B., Vaughan, J. W., Wallach, H., Daumé, H., & Crawford, K. (2018) · arXiv · doi:10.48550/arxiv.1803.09010 · Cited by 2,474
Gebru, T., Morgenstern, J., Vecchione, B., Vaughan, J. W., Wallach, H., III, H. D., & Crawford, K. (2021) · Communications of the ACM · doi:10.1145/3458723 · Cited by 2,469 · cited 2× · see citing works
“This reflection and metadata benefits transparency for the reader, but also benefits the researcher, in a similar vein to that described by Gebru et al. in [21].”
BackgroundMethodRacial disparities in automated speech recognition
Koenecke, A., Nam, A., Lake, E., Nudell, J., Quartey, M., Mengesha, Z., Toups, C., Rickford, J. R., Jurafsky, D., & Goel, S. (2020) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1915768117 · Cited by 765 · see citing works
The Misgendering Machines: Trans/HCI Implications of Automatic Gender Recognition
Keyes, O. (2018) · Proceedings of the ACM on Human-Computer Interaction · doi:10.1145/3274357 · Cited by 505 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
Tang, X., Li, J., & Wu, S. (2026) · Proceedings of the 2026 CHI Conference on Human Factors in Computing Systems · doi:10.1145/3772318.3790405 · Cited by 4 · see citing works
“Other research has uncovered the ambiguity and uncertainty inherent in socially constructed labels, such as gender [85], accent [81], and content toxicity [1, 16].”
- Authors
- Prinos, K., Patwari, N., & Power, C. A.
- Affiliations
- Washington University in St. Louis (US)
- Relational Communities, USA
- Venue
- The 2024 ACM Conference on Fairness, Accountability, and Transparency
- Publisher
- ACM
- Published
- 2024-06-03
- Pages
- 1245-1254
- OpenAlex
- W4399362524
- Cited by
- 12
- References
- 5
- Citations by year
- 2026: 6 · 2025: 5 · 2024: 1
- FWCI
- 3.55
- Type
- conference paper
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- Locations
- ACM Conference on Fairness, Accountability, and Transparency (FAccT) (publishedVersion, cc-by-nc-sa)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Speech Recognition and Synthesis
- Social Sciences › Linguistics and Language › Linguistic Variation and Morphology
- Social Sciences › Psychology › Experimental and Cognitive Psychology › Phonetics and Phonology Research
- Keywords
- accent, dialects, speech recognition, accent bias, discrimination, listener perception, bias mitigation
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
