Self-supervised learning for infant cry analysis

TL;DR

This paper demonstrates that self-supervised learning on unlabeled infant cry data improves detection of neurological injury and cry triggers, reducing the need for extensive labeled datasets in medical cry analysis.

Contribution

It introduces SSL pre-training with SimCLR for infant cry analysis, showing significant performance improvements over supervised methods and reducing labeled data requirements.

Findings

SSL pre-training outperforms supervised pre-training.

SSL-based domain adaptation enhances cry analysis accuracy.

Reduces need for labeled data in clinical cry detection.

Abstract

In this paper, we explore self-supervised learning (SSL) for analyzing a first-of-its-kind database of cry recordings containing clinical indications of more than a thousand newborns. Specifically, we target cry-based detection of neurological injury as well as identification of cry triggers such as pain, hunger, and discomfort. Annotating a large database in the medical setting is expensive and time-consuming, typically requiring the collaboration of several experts over years. Leveraging large amounts of unlabeled audio data to learn useful representations can lower the cost of building robust models and, ultimately, clinical solutions. In this work, we experiment with self-supervised pre-training of a convolutional neural network on large audio datasets. We show that pre-training with SSL contrastive loss (SimCLR) performs significantly better than supervised pre-training for both neuro injury and cry triggers. In addition, we demonstrate further performance gains through SSL-based domain adaptation using unlabeled infant cries. We also show that using such SSL-based pre-training for adaptation to cry sounds decreases the need for labeled data of the overall system.