An Acoustic Segment Model Based Segment Unit Selection Approach to Acoustic Scene Classification with Partial Utterances

TL;DR

This paper introduces a novel sub-utterance unit selection method for acoustic scene classification that improves accuracy by removing low-information segments, leveraging acoustic segment models and a stop word analogy.

Contribution

The paper presents a new framework for segment selection in ASC using acoustic segment models and stop ASM detection, enhancing classification accuracy without data augmentation.

Findings

Accuracy improved from 68% to 72.1% on DCASE 2018 dataset.

Competitive performance without data augmentation or ensemble methods.

Effective removal of low-information segments enhances scene classification.

Abstract

In this paper, we propose a sub-utterance unit selection framework to remove acoustic segments in audio recordings that carry little information for acoustic scene classification (ASC). Our approach is built upon a universal set of acoustic segment units covering the overall acoustic scene space. First, those units are modeled with acoustic segment models (ASMs) used to tokenize acoustic scene utterances into sequences of acoustic segment units. Next, paralleling the idea of stop words in information retrieval, stop ASMs are automatically detected. Finally, acoustic segments associated with the stop ASMs are blocked, because of their low indexing power in retrieval of most acoustic scenes. In contrast to building scene models with whole utterances, the ASM-removed sub-utterances, i.e., acoustic utterances without stop acoustic segments, are then used as inputs to the AlexNet-L back-end for final classification. On the DCASE 2018 dataset, scene classification accuracy increases from 68%, with whole utterances, to 72.1%, with segment selection. This represents a competitive accuracy without any data augmentation, and/or ensemble strategy. Moreover, our approach compares favourably to AlexNet-L with attention.