TY - JOUR

T1 - ERANNs: Efficient residual audio neural networks for audio pattern recognition

AU - Verbitskiy, Sergey

AU - Berikov, Vladimir

AU - Vyshegorodtsev, Viacheslav

N1 - Funding Information: This work was partly supported by NVIDIA Inception Program for AI Startups; FASIE grant 0060584; RFBR grant 19-29-01175. We thank the anonymous reviewers, whose suggestions helped improve the manuscript quality. Publisher Copyright: © 2022 Elsevier B.V.

PY - 2022/9

Y1 - 2022/9

N2 - Audio pattern recognition (APR) is an important research topic and can be applied to several fields related to our lives. Therefore, accurate and efficient APR systems need to be developed as they are useful in real applications. In this paper, we propose a new convolutional neural network (CNN) architecture and a method for improving the inference speed of CNN-based systems for APR tasks. Moreover, using the proposed method, we can improve the performance of our systems, as confirmed in experiments conducted on four audio datasets. In addition, we investigate the impact of data augmentation techniques and transfer learning on the performance of our systems. Our best system achieves a mean average precision (mAP) of 0.450 on the AudioSet dataset. Although this value is less than that of the state-of-the-art system, the proposed system is 7.1x faster and 9.7x smaller. On the ESC-50, UrbanSound8K, and RAVDESS datasets, we obtain state-of-the-art results with accuracies of 0.961, 0.908, and 0.748, respectively. Our system for the ESC-50 dataset is 1.7x faster and 2.3x smaller than the previous best system. For the RAVDESS dataset, our system is 3.3x smaller than the previous best system. We name our systems “Efficient Residual Audio Neural Networks”.

AB - Audio pattern recognition (APR) is an important research topic and can be applied to several fields related to our lives. Therefore, accurate and efficient APR systems need to be developed as they are useful in real applications. In this paper, we propose a new convolutional neural network (CNN) architecture and a method for improving the inference speed of CNN-based systems for APR tasks. Moreover, using the proposed method, we can improve the performance of our systems, as confirmed in experiments conducted on four audio datasets. In addition, we investigate the impact of data augmentation techniques and transfer learning on the performance of our systems. Our best system achieves a mean average precision (mAP) of 0.450 on the AudioSet dataset. Although this value is less than that of the state-of-the-art system, the proposed system is 7.1x faster and 9.7x smaller. On the ESC-50, UrbanSound8K, and RAVDESS datasets, we obtain state-of-the-art results with accuracies of 0.961, 0.908, and 0.748, respectively. Our system for the ESC-50 dataset is 1.7x faster and 2.3x smaller than the previous best system. For the RAVDESS dataset, our system is 3.3x smaller than the previous best system. We name our systems “Efficient Residual Audio Neural Networks”.

KW - Audio pattern recognition

KW - Audio tagging

KW - Residual convolutional neural networks

KW - Sound classification

KW - Transfer learning

UR - http://www.scopus.com/inward/record.url?scp=85134814345&partnerID=8YFLogxK

U2 - 10.1016/j.patrec.2022.07.012

DO - 10.1016/j.patrec.2022.07.012

M3 - Article

AN - SCOPUS:85134814345

VL - 161

SP - 38

EP - 44

JO - Pattern Recognition Letters

JF - Pattern Recognition Letters

SN - 0167-8655

ER -