Automatic Speech Recognition of Disordered Speech: Personalized Models Outperforming Human Listeners on Short Phrases
The DOI matches its Crossref record; 9 of the 9 references it cites were found in a registry or an index.
Abstract
This study evaluated the accuracy of personalized automatic speech recognition (ASR) for recognizing disordered speech from a large cohort of individuals with a wide range of underlying etiologies using an open vocabulary. The performance of these models was benchmarked relative to that of expert human transcribers and two different speaker-independent ASR models trained on typical speech. 432 individuals with self-reported disordered speech recorded at least 300 short phrases using a web-based application. Word error rates (WERs) were estimated for three different ASR models and for human transcribers. Metadata were collected to evaluate the potential impact of participants, atypical speech characteristics, and technical factors on recognition accuracy. Personalized models outperformed human transcribers with median and max recognition accuracy gains of 9% and 80%, respectively. The accuracies of personalized models were high (median WER: 4.6%) and better than those of speaker-independent models (median WER: 31%). The most significant improvements were for the most severely affected speakers. Low signal-to-noise ratio and fewer training utterances were associated with poor word recognition, even for speakers with mild speech impairments. Our results demonstrate the efficacy of personalized ASR models in recognizing a wide range of speech impairments and severities and using an open vocabulary.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
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?
In the citing paper
“For this speaker with severe dysarthria and global aphasia, a speaker-dependent ASR therefore functions as an amplifier of intelligibility, supporting more reliable communication in everyday contexts (17, 18).”
From Case Report: Tailored automatic speech recognition in global aphasia with dysarthria - a single case proof of concept (2026) · cites this paper as [18]
Citations
Cites
SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition
Park, D. S., Chan, W., Zhang, Y., Chiu, C.-C., Zoph, B., Cubuk, E. D., & Le, Q. V. (2019) · Interspeech 2019 · doi:10.21437/interspeech.2019-2680 · Cited by 3,558
Motor Speech Disorders: A Cross-Language Perspective
(2014) · Channel View Publications eBooks · doi:10.2307/jj.26931941 · Cited by 1,260
Robust signal-to-noise ratio estimation based on waveform amplitude distribution analysis
Kim, C., & Stern, R. M. (2008) · Interspeech 2008 · doi:10.21437/interspeech.2008-644 · Cited by 165
Predicting Early Bulbar Decline in Amyotrophic Lateral Sclerosis: A Speech Subsystem Approach
Rong, P., Yunusova, Y., Wang, J., & Green, J. R. (2015) · Behavioural Neurology · doi:10.1155/2015/183027 · Cited by 149
Self-citationLee, J., Hustad, K. C., & Weismer, G. G. (2014) · Journal of Speech Language and Hearing Research · doi:10.1044/2014_jslhr-s-13-0292 · Cited by 120
Shahamiri, S. R., & Salim, S. S. (2014) · Advanced Engineering Informatics · doi:10.1016/j.aei.2014.01.001 · Cited by 84
Ferrier, L., Shane, H., Ballard, H., Carpenter, T., & Benoit, A. (1995) · Augmentative and Alternative Communication · doi:10.1080/07434619512331277289 · Cited by 78
Accuracy of three speech recognition systems: Case study of dysarthric speech
Hux, K., Rankin-Erickson, J., Manasse, N., & Lauritzen, E. (2000) · Augmentative and Alternative Communication · doi:10.1080/07434610012331279044 · Cited by 60
Dysarthric speech recognition using dysarthria-severity-dependent and speaker-adaptive models
Kim, M. J., Yoo, J., & Kim, H. (2013) · Interspeech (USB) · doi:10.21437/interspeech.2013-320 · Cited by 55
Cited by
Items 101–115 of 115
Deep Learning aplicado a reconocimiento y caracterización de habla en usuarios con síndrome de Down
Fernández García, D. (2024) · University of Valladolid Documentary Repository (University of Valladolid)
Ferri-Molla, I. (2023) · RiuNet (Universitat Politècnica de València)
ASR model adaptation to the speech of people with Down syndrome
García, D. F., Cardeñoso-Payo, V., González-Ferreras, C., & Escudero-Mancebo, D. (2024) · UVaDOC UVaDOC University of Valladolid Documentary Repository (University of Valladolid) · doi:10.26342/2024-73-15
Improving Aphasic Communication Using Multimodal AI Systems
Ferri-Molla, I., Linares-Pellicer, J., & Izquierdo-Domenech, J. (2026) · International Journal of Interactive Multimedia and Artificial Intelligence · doi:10.9781/ijimai.2026.2215
Mobile-based speech-to-text and response application
Sangalang, T. K. M., Binalla, M. M. R., Francisco, P. F. Y., Garcia, J. N. M., Salongsongan, S. P., Lopez, A. K. M. D., Viaña, J. X. R., Bautista, J. N., Galay-Limos, J. A., & Rodil, M. V. B. (2026) · International Journal of Research Studies in Educational Technology · doi:10.5861/ijrset.2026.26009
Traduction automatique de la parole vers des pictogrammes
Macaire, C. (2025) · theses.fr (ABES) · doi:10.70675/06c7e6b3z41b7z467cza019zc899470d6306
Lindby, S. B., Jensen, J., Tan, Z., & Østergaard, J. (2026) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp55912.2026.11461245
Ferri-Molla, I. (2026) · RiuNet (Universitat Politècnica de València) · doi:10.4995/thesis/10251/235897
Comparing Human and Automatic Recognition of Dutch Dysarthric Continuous Speech: A Case StudyPreprint
Zhang, Y., de Groot, D., Martínez, J., & Scharenborg, O. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2606.30237
ASR Personalization in Pathological Voice: Accuracy, Efficiency, and Duration Effects
Giaimis, G., Katkov, S., Yaqub, M. A., & Liotta, A. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37867-5_15
Rykova, E. (2026) · UEF eRepo (University of Eastern Finland)
Human-calibrated automated intelligibility evaluation of speech neuroprosthesesPreprint
Srinivasan, A., Wairagkar, M., Wakeb, R., Farias, D., Miller, L. M., Brandman, D. M., & Stavisky, S. D. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.10.27.762196
박규태, & Kim, J. (2026) · Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP) · doi:10.18653/2026.findings-emnlp.409
Nguyen, K., Dinh, T., & Phan, B. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2611.07587
DIY Customized Speech Recognition for a Person with Speech Disabilities
Condado, P. A., & Lobo, F. G. (2026) · arXiv · doi:10.48550/arxiv.2611.11327
Metadata
- Authors
- Green, J. R., MacDonald, R. L., Jiang, P.-P., Cattiau, J., Heywood, R., Cave, R., Seaver, K., Ladewig, M. A., Tobin, J., Brenner, M. P., Nelson, P. C., & Tomanek, K.
- ORCID
- Venue
- Interspeech 2021
- Publisher
- ISCA
- Published
- 2021-08-30
- Pages
- 4778-4782
- ISSN
- 2308-457X (ISSN-L)
- OpenAlex
- W3197712601
- Cited by
- 74
- References
- 9
- Citations by year
- 2026: 5 · 2025: 18 · 2024: 22 · 2023: 12 · 2022: 16 · 2021: 1
- FWCI
- 5.89
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- Interspeech (USB) (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Speech Recognition and Synthesis
- Health Sciences › Health Professions › Speech and Hearing › Voice and Speech Disorders
- Physical Sciences › Computer Science › Signal Processing › Speech and Audio Processing
- Keywords
- speech recognition, disordered speech, short phrases, human listeners
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
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.