From Human Speech to Ocean Signals: Transferring Speech Large Models for Underwater Acoustic Target Recognition

TL;DR

This paper demonstrates that large speech models can be effectively transferred to underwater acoustic target recognition, achieving high accuracy and robustness despite limited labeled data.

Contribution

It introduces UATR-SLM, a framework that adapts speech large models for underwater acoustics, showing promising transferability and performance.

Findings

Achieves over 99% in-domain accuracy

Maintains robustness across variable signal lengths

Reaches 96.67% accuracy in cross-domain tests

Abstract

Underwater acoustic target recognition (UATR) plays a vital role in marine applications but remains challenging due to limited labeled data and the complexity of ocean environments. This paper explores a central question: can speech large models (SLMs), trained on massive human speech corpora, be effectively transferred to underwater acoustics? To investigate this, we propose UATR-SLM, a simple framework that reuses the speech feature pipeline, adapts the SLM as an acoustic encoder, and adds a lightweight classifier.Experiments on the DeepShip and ShipsEar benchmarks show that UATR-SLM achieves over 99% in-domain accuracy, maintains strong robustness across variable signal lengths, and reaches up to 96.67% accuracy in cross-domain evaluation. These results highlight the strong transferability of SLMs to UATR, establishing a promising paradigm for leveraging speech foundation models in underwater acoustics.