Unsupervised adversarial domain adaptation for acoustic scene classification

TL;DR

This paper introduces the first unsupervised adversarial domain adaptation method for acoustic scene classification, significantly improving accuracy on unseen, unlabeled data with minimal impact on labeled data.

Contribution

It presents a novel unsupervised adversarial approach for domain adaptation in acoustic scene classification, addressing mismatched conditions without requiring labels for target data.

Findings

Achieved approximately 10% increase in accuracy on unseen, unlabeled data.

Maintained nearly the same performance on labeled source data.

Demonstrated effectiveness using DCASE 2018 dataset with mismatched recording devices.

Abstract

A general problem in acoustic scene classification task is the mismatched conditions between training and testing data, which significantly reduces the performance of the developed methods on classification accuracy. As a countermeasure, we present the first method of unsupervised adversarial domain adaptation for acoustic scene classification. We employ a model pre-trained on data from one set of conditions and by using data from other set of conditions, we adapt the model in order that its output cannot be used for classifying the set of conditions that input data belong to. We use a freely available dataset from the DCASE 2018 challenge Task 1, subtask B, that contains data from mismatched recording devices. We consider the scenario where the annotations are available for the data recorded from one device, but not for the rest. Our results show that with our model agnostic method we can achieve $\sim 10\%$ increase at the accuracy on an unseen and unlabeled dataset, while keeping almost the same performance on the labeled dataset.