Interpretable Deep Learning Model for the Detection and Reconstruction of Dysarthric Speech

Daniel Korzekwa; Roberto Barra-Chicote; Bożena Kostek; Thomas Drugman; Mateusz Łajszczak

doi:10.21437/interspeech.2019-1206

Interpretable Deep Learning Model for the Detection and Reconstruction of Dysarthric Speech

Abstract

We present a novel deep learning model for the detection and reconstruction of dysarthric speech. We train the model with a multi-task learning technique to jointly solve dysarthria detection and speech reconstruction tasks. The model key feature is a low-dimensional latent space that is meant to encode the properties of dysarthric speech. It is commonly believed that neural networks are black boxes that solve problems but do not provide interpretable outputs. On the contrary, we show that this latent space successfully encodes interpretable characteristics of dysarthria, is effective at detecting dysarthria, and that manipulation of the latent space allows the model to reconstruct healthy speech from dysarthric speech. This work can help patients and speech pathologists to improve their understanding of the condition, lead to more accurate diagnoses and aid in reconstructing healthy speech for afflicted patients.

Citations

1 4

CrossRef
0

Web of Science
2 0

Scopus

Authors (5)

Daniel Korzekwa mgr inż.
Roberto Barra-Chicote prof.
Bożena Kostek prof. dr hab. inż.
Thomas Drugman dr
Mateusz Łajszczak

Cite as

Full text

download paper

downloaded 89 times

Publication version: Accepted or Published Version
License: Copyright (2019 ISCA)

Keywords

DYSARTHRIA DETECTION, SPEECH RECOGNITION, SPEECH SYNTHESIS, INTERPRETABLE DEEP LEARNING MODELS

Details

Category:

Conference activity

Type:

publikacja w wydawnictwie zbiorowym recenzowanym (także w materiałach konferencyjnych)

Language:

English

Publication year:

2019

Bibliographic description:

Korzekwa D., Barra-Chicote R., Kostek B., Drugman T., Łajszczak M.: Interpretable Deep Learning Model for the Detection and Reconstruction of Dysarthric Speech// / : , 2019,

DOI: