SESQA: semi-supervised learning for speech quality assessment

2020/10/01 by Serrà, Joan, Pons, Jordi, Pascual, Santiago · 2 citations
#Audio and Speech Processing (eess.AS) #FOS: Computer and information sciences #FOS: Electrical engineering #Machine Learning (cs.LG) #Sound (cs.SD) #electronic engineering #information engineering

paper · doi:10.48550/arxiv.2010.00368

Abstract

Automatic speech quality assessment is an important, transversal task whose progress is hampered by the scarcity of human annotations, poor generalization to unseen recording conditions, and a lack of flexibility of existing approaches. In this work, we tackle these problems with a semi-supervised learning approach, combining available annotations with programmatically generated data, and using 3 different optimization criteria together with 5 complementary auxiliary tasks. Our results show that such a semi-supervised approach can cut the error of existing methods by more than 36%, while providing additional benefits in terms of reusable features or auxiliary outputs. Improvement is further corroborated with an out-of-sample test showing promising generalization capabilities.

Cited by

• Training a Perceptual Model for Evaluating Auditory Similarity in Music Adversarial Attack
• JSQA: Speech Quality Assessment with Perceptually-Inspired Contrastive Pretraining Based on JND Audio Pairs

Related

• Improving Speech Enhancement with Multi-Metric Supervision from Learned Quality Assessment
• Semi-supervised Learning For Robust Speech Evaluation
• MOS-Bench: Benchmarking Generalization Abilities of Subjective Speech Quality Assessment Models
• Speech Quality Assessment Model Based on Mixture of Experts: System-Level Performance Enhancement and Utterance-Level Challenge Analysis
• Self-Supervised Speech Quality Assessment (S3QA): Leveraging Speech Foundation Models for a Scalable Speech Quality Metric