Adaptive DCTNet for Audio Signal Classification
This work addresses audio signal classification problems, such as bird song and artist identification, with an incremental improvement over existing methods.
The paper tackled audio signal classification by introducing Adaptive DCTNet (A-DCTNet), which adapts to different acoustic scales and better captures low-frequency information than MFSC features. Experimental results showed that A-DCTNet with RNNs achieved state-of-the-art performance in bird song classification and improved artist identification accuracy in music data.
In this paper, we investigate DCTNet for audio signal classification. Its output feature is related to Cohen's class of time-frequency distributions. We introduce the use of adaptive DCTNet (A-DCTNet) for audio signals feature extraction. The A-DCTNet applies the idea of constant-Q transform, with its center frequencies of filterbanks geometrically spaced. The A-DCTNet is adaptive to different acoustic scales, and it can better capture low frequency acoustic information that is sensitive to human audio perception than features such as Mel-frequency spectral coefficients (MFSC). We use features extracted by the A-DCTNet as input for classifiers. Experimental results show that the A-DCTNet and Recurrent Neural Networks (RNN) achieve state-of-the-art performance in bird song classification rate, and improve artist identification accuracy in music data. They demonstrate A-DCTNet's applicability to signal processing problems.