The Efficacy of Self-Supervised Speech Models for Audio Representations

TL;DR

This paper demonstrates that self-supervised speech models, when combined with ensemble techniques, can effectively produce robust audio representations for both speech and non-speech tasks, surpassing existing models in many cases.

Contribution

The authors propose an ensemble framework for SSL speech models, showing its effectiveness across diverse audio datasets and outperforming state-of-the-art models in the HEAR Challenge.

Findings

SSL speech models perform well on non-speech tasks

Ensemble techniques improve representation quality

Framework surpasses state-of-the-art models in HEAR Challenge

Abstract

Self-supervised learning (SSL) speech models, which can serve as powerful upstream models to extract meaningful speech representations, have achieved unprecedented success in speech representation learning. However, their effectiveness on non-speech datasets is relatively less explored. In this work, we propose an ensemble framework, with a combination of ensemble techniques, to fuse SSL speech models' embeddings. Extensive experiments on speech and non-speech audio datasets are conducted to investigate the representation abilities of our ensemble method and its single constituent model. Ablation studies are carried out to evaluate the performances of different ensemble techniques, such as feature averaging and concatenation. All experiments are conducted during NeurIPS 2021 HEAR Challenge as a standard evaluation pipeline provided by competition officials. Results demonstrate SSL speech models' strong abilities on various non-speech tasks, while we also note that they fail to deal with fine-grained music tasks, such as pitch classification and note onset detection. In addition, feature ensemble is shown to have great potential on producing more holistic representations, as our proposed framework generally surpasses state-of-the-art SSL speech/audio models and has superior performance on various datasets compared with other teams in HEAR Challenge. Our code is available at https://github.com/tony10101105/HEAR-2021-NeurIPS-Challenge -- NTU-GURA.