Multi-Task Pseudo-Label Learning for Non-Intrusive Speech Quality Assessment Model

TL;DR

This paper introduces MTQ-Net, a non-intrusive speech quality assessment model that leverages multi-task pseudo-label learning to improve prediction accuracy by combining pseudo-labels from a pretrained model with ground-truth data.

Contribution

The study presents a novel multi-task pseudo-label learning framework for speech quality assessment, demonstrating improved performance over existing SSL-based models.

Findings

MPL outperforms training from scratch and direct knowledge transfer.

Huber loss enhances predictive accuracy.

MTQ-Net achieves higher predictive power than other SSL-based models.

Abstract

This study proposes a multi-task pseudo-label learning (MPL)-based non-intrusive speech quality assessment model called MTQ-Net. MPL consists of two stages: obtaining pseudo-label scores from a pretrained model and performing multi-task learning. The 3QUEST metrics, namely Speech-MOS (S-MOS), Noise-MOS (N-MOS), and General-MOS (G-MOS), are the assessment targets. The pretrained MOSA-Net model is utilized to estimate three pseudo labels: perceptual evaluation of speech quality (PESQ), short-time objective intelligibility (STOI), and speech distortion index (SDI). Multi-task learning is then employed to train MTQ-Net by combining a supervised loss (derived from the difference between the estimated score and the ground-truth label) and a semi-supervised loss (derived from the difference between the estimated score and the pseudo label), where the Huber loss is employed as the loss function. Experimental results first demonstrate the advantages of MPL compared to training a model from scratch and using a direct knowledge transfer mechanism. Second, the benefit of the Huber loss for improving the predictive ability of MTQ-Net is verified. Finally, the MTQ-Net with the MPL approach exhibits higher overall predictive power compared to other SSL-based speech assessment models.