TC-SKNet with GridMask for Low-complexity Classification of Acoustic scene

TL;DR

This paper introduces TC-SKNet with GridMask, a low-complexity CNN model for acoustic scene classification that adapts to variable speech lengths and uses AutoML for optimal structure, achieving competitive accuracy with fewer parameters.

Contribution

It proposes a novel combination of Selective Kernel Network with Temporal-Convolution and GridMask data augmentation, optimized via AutoML, for efficient acoustic scene classification.

Findings

Achieved 59.87% accuracy with only 20.9K parameters.

GridMask outperforms spectrum augmentation in performance gains.

AutoML effectively optimizes model structure and hyperparameters.

Abstract

Convolution neural networks (CNNs) have good performance in low-complexity classification tasks such as acoustic scene classifications (ASCs). However, there are few studies on the relationship between the length of target speech and the size of the convolution kernels. In this paper, we combine Selective Kernel Network with Temporal-Convolution (TC-SKNet) to adjust the receptive field of convolution kernels to solve the problem of variable length of target voice while keeping low-complexity. GridMask is a data augmentation strategy by masking part of the raw data or feature area. It can enhance the generalization of the model as the role of dropout. In our experiments, the performance gain brought by GridMask is stronger than spectrum augmentation in ASCs. Finally, we adopt AutoML to search best structure of TC-SKNet and hyperparameters of GridMask for improving the classification performance. As a result, a peak accuracy of 59.87% TC-SKNet is equivalent to that of SOTA, but the parameters only use 20.9 K.