SpecMix : A Mixed Sample Data Augmentation method for Training withTime-Frequency Domain Features
This work addresses the need for effective data augmentation in audio processing, particularly for time-frequency features, but it is incremental as it adapts existing mixed sample strategies to a specific domain.
The authors tackled the problem of improving model performance on audio tasks by proposing SpecMix, a mixed sample data augmentation method for time-frequency domain features, which achieved a maximum performance improvement of 2.7% on tasks like acoustic scene classification and sound event classification.
A mixed sample data augmentation strategy is proposed to enhance the performance of models on audio scene classification, sound event classification, and speech enhancement tasks. While there have been several augmentation methods shown to be effective in improving image classification performance, their efficacy toward time-frequency domain features of audio is not assured. We propose a novel audio data augmentation approach named "Specmix" specifically designed for dealing with time-frequency domain features. The augmentation method consists of mixing two different data samples by applying time-frequency masks effective in preserving the spectral correlation of each audio sample. Our experiments on acoustic scene classification, sound event classification, and speech enhancement tasks show that the proposed Specmix improves the performance of various neural network architectures by a maximum of 2.7%.