# Generalizability of Predictive and Generative Speech Enhancement Models to Pathological Speakers

arXiv: https://arxiv.org/abs/2509.18890 · HTML: https://mlsanity.com/paper/2509.18890/

## Key facts

- Cited by 1 papers in the map (data through 19 Oct 2026).
- Placed in ASR evaluation, bias and atypical speech, one of 11 directions in Speech and audio language models.
- Builds on 19 of its 39 references that are in the map.
- First posted on arXiv 23 Sep 2025.

## Placed in

- Speech and audio language models / [ASR evaluation, bias and atypical speech](https://mlsanity.com/speech/asr-evaluation-bias-and-atypical-speech/)

## Builds on (most cited first, 19 of 19)

- [Score-Based Generative Modeling through Stochastic Differential Equations](https://mlsanity.com/paper/2011.13456/) (2020)
- [Generative Modeling by Estimating Gradients of the Data Distribution](https://mlsanity.com/paper/1907.05600/) (2019)
- [SDR - half-baked or well done?](https://mlsanity.com/paper/1811.02508/) (2018)
- [Supervised Speech Separation Based on Deep Learning: An Overview](https://mlsanity.com/paper/1708.07524/) (2017)
- [SEGAN: Speech Enhancement Generative Adversarial Network](https://mlsanity.com/paper/1703.09452/) (2017)
- [Speech Enhancement and Dereverberation with Diffusion-based Generative Models](https://mlsanity.com/paper/2208.05830/) (2022)
- [Conditional Diffusion Probabilistic Model for Speech Enhancement](https://mlsanity.com/paper/2202.05256/) (2022)
- [StoRM: A Diffusion-based Stochastic Regeneration Model for Speech Enhancement and Dereverberation](https://mlsanity.com/paper/2212.11851/) (2022)
- [A Study on Speech Enhancement Based on Diffusion Probabilistic Model](https://mlsanity.com/paper/2107.11876/) (2021)
- [Improving noise robust automatic speech recognition with single-channel time-domain enhancement network](https://mlsanity.com/paper/2003.03998/) (2020)
- [URGENT Challenge: Universality, Robustness, and Generalizability For Speech Enhancement](https://mlsanity.com/paper/2406.04660/) (2024)
- [A consolidated view of loss functions for supervised deep learning-based speech enhancement](https://mlsanity.com/paper/2009.12286/) (2020)
- [Schr"odinger Bridge for Generative Speech Enhancement](https://mlsanity.com/paper/2407.16074/) (2024)
- [Diffusion Models for Audio Restoration](https://mlsanity.com/paper/2402.09821/) (2024)
- [Schrodinger Bridges Beat Diffusion Models on Text-to-Speech Synthesis](https://mlsanity.com/paper/2312.03491/) (2023)
- [Investigating Training Objectives for Generative Speech Enhancement](https://mlsanity.com/paper/2409.10753/) (2024)
- [Assessing the Generalization Gap of Learning-Based Speech Enhancement Systems in Noisy and Reverberant Environments](https://mlsanity.com/paper/2309.06183/) (2023)
- [Diffusion-Based Speech Enhancement in Matched and Mismatched Conditions Using a Heun-Based Sampler](https://mlsanity.com/paper/2312.02683/) (2023)
- [Variational Autoencoder for Personalized Pathological Speech Enhancement](https://mlsanity.com/paper/2503.14036/) (2025)

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

- [Data Augmentation for Pathological Speech Enhancement](https://mlsanity.com/paper/2602.14671/) (2026)

## Abstract (arXiv)

State of the art speech enhancement (SE) models achieve strong performance on neurotypical speech, but their effectiveness is substantially reduced for pathological speech. In this paper, we investigate strategies to address this gap for both predictive and generative SE models, including i) training models from scratch using pathological data, ii) finetuning models pretrained on neurotypical speech with additional data from pathological speakers, and iii) speaker specific personalization using only data from the individual pathological test speaker. Our results show that, despite the limited size of pathological speech datasets, SE models can be successfully trained or finetuned on such data. Finetuning models with data from several pathological speakers yields the largest performance improvements, while speaker specific personalization is less effective, likely due to the small amount of data available per speaker. These findings highlight the challenges and potential strategies for improving SE performance for pathological speakers.