An Automatic Sound Classification
Framework with Non-volatile Memory
Jibin Wu, Yansong Chua, Malu Zhang, Haizhou Li, and Kay Chen Tan
Abstract Environmental sounds form part of our daily life. With the advancement
of deep learning models and the abundance of training data, the performance of automatic sound classification (ASC) systems has improved significantly in recent years.
However, the high computational cost, hence high power consumption, remains a
major hurdle for large-scale implementation of ASC systems on mobile and wearable
devices. Motivated by the observations that humans are highly effective and consume
little power whilst analyzing complex audio scenes, a biologically plausible ASC
framework is introduced, namely SOM-SNN. The emerging dense crossbar array of
non-volatile memory (NVM) devices have been recognized as a promising approach
to emulate such distributed, massively-parallel and densely connected neuromorphic
computing systems. This chapter presents the general structure of this framework for
sound event and speech recognition, demonstrating attractive computational benefits
and suitableness with an NVM implementation.
1 Introduction
Automatic sound classification (ASC) generally refers to the automatic identification
of the ambient sounds in the environment. Environmental sounds, complementary to
visual cues, provide a great amount of information about our surrounding environment and is an essential part of our daily life. ASC technologies enable a wide range
of applications such as content-based sound classification and retrieval [1], audio
surveillance [2], sound event classification [3] and disease diagnosis [4].
J. Wu (B) · M. Zhang · H. Li
Department of Electrical and Computer Engineering, National University of Singapore,
Singapore, Singapore
e-mail: jibin.wu@u.nus.edu
Y. Chua
Institute for Infocomm Research, A*STAR, Singapore, Singapore
K. C. Tan
Department of Computer Science, City University of Hong Kong, Hong Kong, China
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021
W. S. Lew et al. (eds.), Emerging Non-volatile Memory Technologies,
https://doi.org/10.1007/978-981-15-6912-8_13
415
Framework with Non-volatile Memory
Jibin Wu, Yansong Chua, Malu Zhang, Haizhou Li, and Kay Chen Tan
Abstract Environmental sounds form part of our daily life. With the advancement
of deep learning models and the abundance of training data, the performance of automatic sound classification (ASC) systems has improved significantly in recent years.
However, the high computational cost, hence high power consumption, remains a
major hurdle for large-scale implementation of ASC systems on mobile and wearable
devices. Motivated by the observations that humans are highly effective and consume
little power whilst analyzing complex audio scenes, a biologically plausible ASC
framework is introduced, namely SOM-SNN. The emerging dense crossbar array of
non-volatile memory (NVM) devices have been recognized as a promising approach
to emulate such distributed, massively-parallel and densely connected neuromorphic
computing systems. This chapter presents the general structure of this framework for
sound event and speech recognition, demonstrating attractive computational benefits
and suitableness with an NVM implementation.
1 Introduction
Automatic sound classification (ASC) generally refers to the automatic identification
of the ambient sounds in the environment. Environmental sounds, complementary to
visual cues, provide a great amount of information about our surrounding environment and is an essential part of our daily life. ASC technologies enable a wide range
of applications such as content-based sound classification and retrieval [1], audio
surveillance [2], sound event classification [3] and disease diagnosis [4].
J. Wu (B) · M. Zhang · H. Li
Department of Electrical and Computer Engineering, National University of Singapore,
Singapore, Singapore
e-mail: jibin.wu@u.nus.edu
Y. Chua
Institute for Infocomm Research, A*STAR, Singapore, Singapore
K. C. Tan
Department of Computer Science, City University of Hong Kong, Hong Kong, China
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021
W. S. Lew et al. (eds.), Emerging Non-volatile Memory Technologies,
https://doi.org/10.1007/978-981-15-6912-8_13
415