Exploring Pathological Speech Quality Assessment with ASR-Powered Wav2Vec2 in Data-Scarce Context
Preprint
Abstract
Automatic speech quality assessment has raised more attention as an alternative or support to traditional perceptual clinical evaluation. However, most research so far only gains good results on simple tasks such as binary classification, largely due to data scarcity. To deal with this challenge, current works tend to segment patients' audio files into many samples to augment the datasets. Nevertheless, this approach has limitations, as it indirectly relates overall audio scores to individual segments. This paper introduces a novel approach where the system learns at the audio level instead of segments despite data scarcity. This paper proposes to use the pre-trained Wav2Vec2 architecture for both SSL, and ASR as feature extractor in speech assessment. Carried out on the HNC dataset, our ASR-driven approach established a new baseline compared with other approaches, obtaining average $MSE=0.73$ and $MSE=1.15$ for the prediction of intelligibility and severity scores respectively, using only 95 training samples. It shows that the ASR based Wav2Vec2 model brings the best results and may indicate a strong correlation between ASR and speech quality assessment. We also measure its ability on variable segment durations and speech content, exploring factors influencing its decision.
Keywords
All Related Versions
This publication has 18 references indexed in Scilit:
- Can We Use Speaker Embeddings On Spontaneous Speech Obtained From Medical Conversations To Predict Intelligibility?2023
- Utilizing Wav2Vec In Database-Independent Voice Disorder Detection2023
- Exploring Self-Supervised Pre-Trained ASR Models for Dysarthric and Elderly Speech Recognition2023
- Evidence of Vocal Tract Articulation in Self-Supervised Learning of Speech2023
- Self-Supervised Speech Representation Learning: A ReviewIEEE Journal of Selected Topics in Signal Processing, 2022
- Layer-Wise Analysis of a Self-Supervised Speech Representation Model2021
- C2SI corpus: a database of speech disorder productions to assess intelligibility and quality of life in head and neck cancersLanguage Resources and Evaluation, 2020
- How to manage sound, physiological and clinical data of 2500 dysphonic and dysarthric speakers?Speech Communication, 2012
- Speech technology‐based assessment of phoneme intelligibility in dysarthriaInternational Journal of Language & Communication Disorders, 2009
- A modern approach to dysarthria classification2004