Conditional independence for pretext task selection in Self-supervised speech representation learning

TL;DR

This paper introduces a theoretical framework based on conditional independence to select effective pseudo-labels for self-supervised speech learning, reducing computational costs and improving downstream task performance prediction.

Contribution

It proposes a novel, training-free estimator for pseudo-label utility grounded in conditional independence theory, specifically tailored for speech representation learning.

Findings

Strong correlation between utility estimates and downstream performance

Effective pseudo-label selection reduces computational costs

Validated on speaker recognition and speech recognition tasks

Abstract

Through solving pretext tasks, self-supervised learning (SSL) leverages unlabeled data to extract useful latent representations replacing traditional input features in the downstream task. A common pretext task consists in pretraining a SSL model on pseudo-labels derived from the original signal. This technique is particularly relevant for speech data where various meaningful signal processing features may serve as pseudo-labels. However, the process of selecting pseudo-labels, for speech or other types of data, remains mostly unexplored and currently relies on observing the results on the final downstream task. Nevertheless, this methodology is not sustainable at scale due to substantial computational (hence carbon) costs. Thus, this paper introduces a practical and theoretical framework to select relevant pseudo-labels with respect to a given downstream task. More precisely, we propose a functional estimator of the pseudo-label utility grounded in the conditional independence theory, which does not require any training. The experiments conducted on speaker recognition and automatic speech recognition validate our estimator, showing a significant correlation between the performance observed on the downstream task and the utility estimates obtained with our approach, facilitating the prospection of relevant pseudo-labels for self-supervised speech representation learning.