mlsanityData to 14 Oct 2026

Personalizing ASR for Dysarthric and Accented Speech with Limited Data

arXiv 1907.13511 ↗First posted 31 Jul 2019cs.CL cs.LG cs.SDcited by 45

Not on the map. Why?

Abstract (arXiv)

Automatic speech recognition (ASR) systems have dramatically improved over the last few years. ASR systems are most often trained from 'typical' speech, which means that underrepresented groups don't experience the same level of improvement. In this paper, we present and evaluate finetuning techniques to improve ASR for users with non-standard speech. We focus on two types of non-standard speech: speech from people with amyotrophic lateral sclerosis (ALS) and accented speech. We train personalized models that achieve 62% and 35% relative WER improvement on these two groups, bringing the absolute WER for ALS speakers, on a test set of message bank phrases, down to 10% for mild dysarthria and 20% for more serious dysarthria. We show that 71% of the improvement comes from only 5 minutes of training data. Finetuning a particular subset of layers (with many fewer parameters) often gives better results than finetuning the entire model. This is the first step towards building state of the art ASR models for dysarthric speech.

Key facts. Cited by 45 papers in the map (data through 14 Oct 2026). First posted on arXiv 31 Jul 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 10; 2022: 8; 2023: 6; 2024: 8; 2025: 6; 2026: 7.

45 papers.

Sort:

By area:

  1. 2021Universal Paralinguistic Speech Representations Using Self-Supervised Conformerscited by 32
  2. 2021Residual Adapters for Parameter-Efficient ASR Adaptation to Atypical and Accented Speechcited by 31
  3. 2022Speaker Adaptation Using Spectro-Temporal Deep Features for Dysarthric and Elderly Speech Recognitioncited by 19
  4. 2021Widening Access to Applied Machine Learning with TinyMLcited by 17
  5. 2022Personalized Adversarial Data Augmentation for Dysarthric and Elderly Speech Recognitioncited by 16
  6. 2024Lost in Transcription: Identifying and Quantifying the Accuracy Biases of Automatic Speech Recognition Systems Against Disfluent Speechcited by 15
  7. 2021Personalized Automatic Speech Recognition Trained on Small Disordered Speech Datasetscited by 14
  8. 2023Exploring Self-supervised Pre-trained ASR Models For Dysarthric and Elderly Speech Recognitioncited by 14
  9. 2022Spectro-Temporal Deep Features for Disordered Speech Assessment and Recognitioncited by 13
  10. 2021Accented Speech Recognition: A Surveycited by 12
  11. 2024Self-supervised ASR Models and Features For Dysarthric and Elderly Speech Recognitioncited by 9
  12. 2022An analysis of degenerating speech due to progressive dysarthria on ASR performancecited by 6
33 more
  1. 2021On-Device Personalization of Automatic Speech Recognition Models for Disordered Speechcited by 5
  2. 2021Comparing Supervised Models And Learned Speech Representations For Classifying Intelligibility Of Disordered Speech On Selected Phrasescited by 5
  3. 2022On-the-Fly Feature Based Rapid Speaker Adaptation for Dysarthric and Elderly Speech Recognitioncited by 5
  4. 2023Hyper-parameter Adaptation of Conformer ASR Systems for Elderly and Dysarthric Speech Recognitioncited by 4
  5. 2023Replay to Remember: Continual Layer-Specific Fine-tuning for German Speech Recognitioncited by 4
  6. 2024Perceiver-Prompt: Flexible Speaker Adaptation in Whisper for Chinese Disordered Speech Recognitioncited by 4
  7. 2024Enhancing Dysarthric Speech Recognition for Unseen Speakers via Prototype-Based Adaptationcited by 4
  8. 2025Adapting Foundation Speech Recognition Models to Impaired Speech: A Semantic Re-chaining Approach for Personalization of German Speechcited by 4
  9. 2025Variational Low-Rank Adaptation for Personalized Impaired Speech Recognitioncited by 4
  10. 2023Multi-pass Training and Cross-information Fusion for Low-resource End-to-end Accented Speech Recognitioncited by 3
  11. 2021Error-driven Fixed-Budget ASR Personalization for Accented Speakerscited by 2
  12. 2022Exploring and Evaluating Personalized Models for Code Generationcited by 2
  13. 2023Latent Phrase Matching for Dysarthric Speechcited by 2
  14. 2024Hypernetworks for Personalizing ASR to Atypical Speechcited by 2
  15. 2024Homogeneous Speaker Features for On-the-Fly Dysarthric and Elderly Speaker Adaptationcited by 2
  16. 2025The Sound of Syntax: Finetuning and Comprehensive Evaluation of Language Models for Speech Pathologycited by 2
  17. 2022Improving Voice Trigger Detection with Metric Learningcited by 1
  18. 2025Large Language Model Data Generation for Enhanced Intent Recognition in German Speechcited by 1
  19. 2025Idiosyncratic Versus Normative Modeling of Atypical Speech Recognition: Dysarthric Case Studiescited by 1
  20. 2025Data-Efficient ASR Personalization for Non-Normative Speech Using an Uncertainty-Based Phoneme Difficulty Score for Guided Samplingcited by 1
  21. 2026Mixture-of-Experts with Intermediate CTC Supervision for Accented Speech Recognitioncited by 1
  22. 2026Adapting Foundation ASR Models to Dysarthric Speech: A Case Studycited by 1
  23. 2021Optimal Transport-based Adaptation in Dysarthric Speech Taskscited by 0
  24. 2021Controllable Response Generation for Assistive Use-casescited by 0
  25. 2022Interpretable Dysarthric Speaker Adaptation based on Optimal-Transportcited by 0
  26. 2023The unreasonable effectiveness of AI CADe polyp detectors to generalize to new countriescited by 0
  27. 2024Predicting Generalization of AI Colonoscopy Models to Unseen Datacited by 0
  28. 2024An Efficient Self-Learning Framework For Interactive Spoken Dialog Systemscited by 0
  29. 2026Unheard in the Digital Age: Rethinking AI Bias and Speech Diversitycited by 0
  30. 2026Multimodal Consistency-Guided Reference-Free Data Selection for ASR Accent Adaptationcited by 0
  31. 2026Comparing Human and Automatic Recognition of Dutch Dysarthric Continuous Speech: A Case Studycited by 0
  32. 2026Analyzing Speech Condition Effects in Dysarthric ASR: A Layer-wise Probing Studycited by 0
  33. 2026Choosing a PEFT Variant for Per-Patient Dysarthric ASR: A Single-Speaker Case Study on Two ASR Basescited by 0