6 papers
Rethinking Batch Composition: Promoting Feature Diversity for Speech Enhancement
Mingchi Hou, Ina Kodrasi
Speech enhancement systems are commonly trained using randomly sampled mini-batches, which may contain limited diversity in intrinsic acoustic characteristics of the clean speech s…
Multiclass Speech Classification Under Noise Disparity
Mahdi Amiri, Sayantan Biswas, Mingchi Hou +2
We investigate noise disparity in multi-class speech classification tasks and develop a strategy to prevent classifiers from exploiting class-dependent noise characteristics. Build…
Data Augmentation for Pathological Speech Enhancement
Mingchi Hou, Enno Hermann, Ina Kodrasi
The performance of state-of-the-art speech enhancement (SE) models considerably degrades for pathological speech due to atypical acoustic characteristics and limited data availabil…
Generalizability of Predictive and Generative Speech Enhancement Models to Pathological Speakers
Mingchi Hou, Ante Jukic, Ina Kodrasi
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…
Influence of Clean Speech Characteristics on Speech Enhancement Performance
Mingchi Hou, Ina Kodrasi
Speech enhancement (SE) performance is known to depend on noise characteristics and signal to noise ratio (SNR), yet intrinsic properties of the clean speech signal itself remain a…
Variational Autoencoder for Personalized Pathological Speech Enhancement
Mingchi Hou, Ina Kodrasi
The generalizability of speech enhancement (SE) models across speaker conditions remains largely unexplored, despite its critical importance for broader applicability. This paper i…