# Idiosyncratic Versus Normative Modeling of Atypical Speech Recognition: Dysarthric Case Studies

arXiv: https://arxiv.org/abs/2509.16718 · HTML: https://mlsanity.com/paper/2509.16718/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 42 references that are in the map.
- First posted on arXiv 20 Sep 2025.

## Builds on (most cited first, 18 of 18)

- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations](https://mlsanity.com/paper/2006.11477/) (2020)
- [Personalizing ASR for Dysarthric and Accented Speech with Limited Data](https://mlsanity.com/paper/1907.13511/) (2019)
- [Instruction-tuning Aligns LLMs to the Human Brain](https://mlsanity.com/paper/2312.00575/) (2023)
- [Residual Adapters for Parameter-Efficient ASR Adaptation to Atypical and Accented Speech](https://mlsanity.com/paper/2109.06952/) (2021)
- [Speaker adaptation for Wav2vec2 based dysarthric ASR](https://mlsanity.com/paper/2204.00770/) (2022)
- [Improved End-to-End Dysarthric Speech Recognition via Meta-learning Based Model Re-initialization](https://mlsanity.com/paper/2011.01686/) (2020)
- [EasyCall corpus: a dysarthric speech dataset](https://mlsanity.com/paper/2104.02542/) (2021)
- [Empirical Evaluation of Pre-trained Transformers for Human-Level NLP: The Role of Sample Size and Dimensionality](https://mlsanity.com/paper/2105.03484/) (2021)
- [Investigating Self-supervised Pretraining Frameworks for Pathological Speech Recognition](https://mlsanity.com/paper/2203.15431/) (2022)
- [Wav2vec-based Detection and Severity Level Classification of Dysarthria from Speech](https://mlsanity.com/paper/2309.14107/) (2023)
- [Self-supervised ASR Models and Features For Dysarthric and Elderly Speech Recognition](https://mlsanity.com/paper/2407.13782/) (2024)
- [An analysis of degenerating speech due to progressive dysarthria on ASR performance](https://mlsanity.com/paper/2211.00089/) (2022)
- [Parameter-efficient Dysarthric Speech Recognition Using Adapter Fusion and Householder Transformation](https://mlsanity.com/paper/2306.07090/) (2023)
- [On-Device Personalization of Automatic Speech Recognition Models for Disordered Speech](https://mlsanity.com/paper/2106.10259/) (2021)
- [Exploring ASR-Based Wav2Vec2 for Automated Speech Disorder Assessment: Insights and Analysis](https://mlsanity.com/paper/2410.08250/) (2024)
- [Multimodal Representation Loss Between Timed Text and Audio for Regularized Speech Separation](https://mlsanity.com/paper/2406.08328/) (2024)

## Built on by (most cited first, 1 of 1)

- [Huntington Disease Automatic Speech Recognition with Biomarker Supervision](https://mlsanity.com/paper/2603.11168/) (2026)

## Abstract (arXiv)

State-of-the-art automatic speech recognition (ASR) models like Whisper, perform poorly on atypical speech, such as that produced by individuals with dysarthria. Past works for atypical speech have mostly investigated fully personalized (or idiosyncratic) models, but modeling strategies that can both generalize and handle idiosyncracy could be more effective for capturing atypical speech. To investigate this, we compare four strategies: (a) textitnormative models trained on typical speech (no personalization), (b) textitidiosyncratic models completely personalized to individuals, (c) textitdysarthric-normative models trained on other dysarthric speakers, and (d) textitdysarthric-idiosyncratic models which combine strategies by first modeling normative patterns before adapting to individual speech. In this case study, we find the dysarthric-idiosyncratic model performs better than idiosyncratic approach while requiring less than half as much personalized data (36.43 WER with 128 train size vs 36.99 with 256). Further, we found that tuning the speech encoder alone (as opposed to the LM decoder) yielded the best results reducing word error rate from 71% to 32% on average. Our findings highlight the value of leveraging both normative (cross-speaker) and idiosyncratic (speaker-specific) patterns to improve ASR for underrepresented speech populations.