LHGNN: Local-Higher Order Graph Neural Networks For Audio Classification and Tagging

TL;DR

This paper introduces LHGNN, a graph neural network that captures higher-order audio relationships using local neighborhoods and fuzzy clustering, outperforming transformers in audio classification tasks with fewer parameters.

Contribution

The work presents a novel graph neural network model that integrates local and higher-order information for improved audio classification, especially without extensive pretraining.

Findings

LHGNN outperforms transformer models on three audio datasets.

LHGNN operates with significantly fewer parameters.

LHGNN is effective without ImageNet pretraining.

Abstract

Transformers have set new benchmarks in audio processing tasks, leveraging self-attention mechanisms to capture complex patterns and dependencies within audio data. However, their focus on pairwise interactions limits their ability to process the higher-order relations essential for identifying distinct audio objects. To address this limitation, this work introduces the Local- Higher Order Graph Neural Network (LHGNN), a graph based model that enhances feature understanding by integrating local neighbourhood information with higher-order data from Fuzzy C-Means clusters, thereby capturing a broader spectrum of audio relationships. Evaluation of the model on three publicly available audio datasets shows that it outperforms Transformer-based models across all benchmarks while operating with substantially fewer parameters. Moreover, LHGNN demonstrates a distinct advantage in scenarios lacking ImageNet pretraining, establishing its effectiveness and efficiency in environments where extensive pretraining data is unavailable.