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

## TL;DR

This paper introduces an interpretable deep learning model that detects and reconstructs dysarthric speech by leveraging a low-dimensional latent space, improving detection accuracy and speech fluency.

## Contribution

It presents a novel encoder-decoder architecture that factorizes speech into an interpretable latent space, enhancing dysarthria detection and speech reconstruction capabilities.

## Key findings

- Latent space conveys interpretable dysarthria characteristics.
- Model generates speech with improved fluency.
- Higher detection accuracy for dysarthria.

## Abstract

This paper proposed a novel approach for the detection and reconstruction of dysarthric speech. The encoder-decoder model factorizes speech into a low-dimensional latent space and encoding of the input text. We showed that the latent space conveys interpretable characteristics of dysarthria, such as intelligibility and fluency of speech. MUSHRA perceptual test demonstrated that the adaptation of the latent space let the model generate speech of improved fluency. The multi-task supervised approach for predicting both the probability of dysarthric speech and the mel-spectrogram helps improve the detection of dysarthria with higher accuracy. This is thanks to a low-dimensional latent space of the auto-encoder as opposed to directly predicting dysarthria from a highly dimensional mel-spectrogram.