Unsupervised Domain Adaptation for Acoustic Scene Classification Using Band-Wise Statistics Matching
This work addresses domain shift issues in acoustic scene classification, which is important for real-world audio applications, but it is incremental as it builds on existing unsupervised adaptation techniques.
The paper tackles the problem of acoustic scene classification performance degradation due to mismatches between training and test data distributions, such as different recording devices, by proposing an unsupervised domain adaptation method that aligns band-wise statistics, resulting in improved classification accuracy on the DCASE 2018 dataset compared to state-of-the-art methods.
The performance of machine learning algorithms is known to be negatively affected by possible mismatches between training (source) and test (target) data distributions. In fact, this problem emerges whenever an acoustic scene classification system which has been trained on data recorded by a given device is applied to samples acquired under different acoustic conditions or captured by mismatched recording devices. To address this issue, we propose an unsupervised domain adaptation method that consists of aligning the first- and second-order sample statistics of each frequency band of target-domain acoustic scenes to the ones of the source-domain training dataset. This model-agnostic approach is devised to adapt audio samples from unseen devices before they are fed to a pre-trained classifier, thus avoiding any further learning phase. Using the DCASE 2018 Task 1-B development dataset, we show that the proposed method outperforms the state-of-the-art unsupervised methods found in the literature in terms of both source- and target-domain classification accuracy.