Voice keyword recognition based on spiking convolutional neural network for human-machine interface
In this paper, a spiking convolutional neural network (SCNN) model for voice keyword recognition is presented. The model consists of an input pre-processing layer, a spiking neural network (SNN) layer with build-in filter bank and the convolutional neural network (CNN) layers. A 16-channel infinite...
| Những tác giả chính: | , , , |
|---|---|
| Tác giả khác: | |
| Định dạng: | Conference Paper |
| Ngôn ngữ: | English |
| Được phát hành: |
2024
|
| Những chủ đề: | |
| Truy cập trực tuyến: | https://hdl.handle.net/10356/179088 https://ieeexplore.ieee.org/abstract/document/9081859 |
| _version_ | 1826122571999870976 |
|---|---|
| author | Hu, Jinhai Goh, Wang Ling Zhang, Zhongyi Gao, Yuan |
| author2 | School of Electrical and Electronic Engineering |
| author_facet | School of Electrical and Electronic Engineering Hu, Jinhai Goh, Wang Ling Zhang, Zhongyi Gao, Yuan |
| author_sort | Hu, Jinhai |
| collection | NTU |
| description | In this paper, a spiking convolutional neural network (SCNN) model for voice keyword recognition is presented. The model consists of an input pre-processing layer, a spiking neural network (SNN) layer with build-in filter bank and the convolutional neural network (CNN) layers. A 16-channel infinite impulse response (IIR) filter bank with energy detector extracts power from the voice signal band and converts it to spikes via the SNN layer. The spiking rate in a defined time window is used as the inputs to the following CNN layers for classification. The network is trained using a voice digit dataset, while the weights of the convolutional layers are adjusted through the training of spike-integration results obtained from the spiking layer. This model has been implemented for voice keyword recognition and achieved 96.0 % accuracy. The combination of SNN and CNN reduces the overall number of layer and neuron in the system without compromise in classification accuracy. It is suitable for low-power hardware implementation in edge devices for human machine interface (HMI) applications. |
| first_indexed | 2024-10-01T05:50:40Z |
| format | Conference Paper |
| id | ntu-10356/179088 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T05:50:40Z |
| publishDate | 2024 |
| record_format | dspace |
| spelling | ntu-10356/1790882024-07-19T15:39:03Z Voice keyword recognition based on spiking convolutional neural network for human-machine interface Hu, Jinhai Goh, Wang Ling Zhang, Zhongyi Gao, Yuan School of Electrical and Electronic Engineering 2020 3rd International Conference on Intelligent Autonomous Systems (ICoIAS) Institute of Microelectronics, A*STAR Engineering Spiking convolutional neural networks Voice keyword recognition In this paper, a spiking convolutional neural network (SCNN) model for voice keyword recognition is presented. The model consists of an input pre-processing layer, a spiking neural network (SNN) layer with build-in filter bank and the convolutional neural network (CNN) layers. A 16-channel infinite impulse response (IIR) filter bank with energy detector extracts power from the voice signal band and converts it to spikes via the SNN layer. The spiking rate in a defined time window is used as the inputs to the following CNN layers for classification. The network is trained using a voice digit dataset, while the weights of the convolutional layers are adjusted through the training of spike-integration results obtained from the spiking layer. This model has been implemented for voice keyword recognition and achieved 96.0 % accuracy. The combination of SNN and CNN reduces the overall number of layer and neuron in the system without compromise in classification accuracy. It is suitable for low-power hardware implementation in edge devices for human machine interface (HMI) applications. Agency for Science, Technology and Research (A*STAR) Submitted/Accepted version This work is funded by A*STAR (Agency for Science, Technology and Research), Singapore under RIE2020 Advanced Manufacturing and Engineering (AME) Programmatic Grant (A18A4b0055). 2024-07-19T05:04:09Z 2024-07-19T05:04:09Z 2020 Conference Paper Hu, J., Goh, W. L., Zhang, Z. & Gao, Y. (2020). Voice keyword recognition based on spiking convolutional neural network for human-machine interface. 2020 3rd International Conference on Intelligent Autonomous Systems (ICoIAS), 77-82. https://dx.doi.org/10.1109/ICoIAS49312.2020.9081859 978-1-7281-6078-8 https://hdl.handle.net/10356/179088 10.1109/ICoIAS49312.2020.9081859 https://ieeexplore.ieee.org/abstract/document/9081859 77 82 en A18A4b0055 © 2020 IEEE. All rights reserved. This article may be downloaded for personal use only. Any other use requires prior permission of the copyright holder. The Version of Record is available online at http://doi.org/10.1109/ICoIAS49312.2020.9081859. application/pdf |
| spellingShingle | Engineering Spiking convolutional neural networks Voice keyword recognition Hu, Jinhai Goh, Wang Ling Zhang, Zhongyi Gao, Yuan Voice keyword recognition based on spiking convolutional neural network for human-machine interface |
| title | Voice keyword recognition based on spiking convolutional neural network for human-machine interface |
| title_full | Voice keyword recognition based on spiking convolutional neural network for human-machine interface |
| title_fullStr | Voice keyword recognition based on spiking convolutional neural network for human-machine interface |
| title_full_unstemmed | Voice keyword recognition based on spiking convolutional neural network for human-machine interface |
| title_short | Voice keyword recognition based on spiking convolutional neural network for human-machine interface |
| title_sort | voice keyword recognition based on spiking convolutional neural network for human machine interface |
| topic | Engineering Spiking convolutional neural networks Voice keyword recognition |
| url | https://hdl.handle.net/10356/179088 https://ieeexplore.ieee.org/abstract/document/9081859 |
| work_keys_str_mv | AT hujinhai voicekeywordrecognitionbasedonspikingconvolutionalneuralnetworkforhumanmachineinterface AT gohwangling voicekeywordrecognitionbasedonspikingconvolutionalneuralnetworkforhumanmachineinterface AT zhangzhongyi voicekeywordrecognitionbasedonspikingconvolutionalneuralnetworkforhumanmachineinterface AT gaoyuan voicekeywordrecognitionbasedonspikingconvolutionalneuralnetworkforhumanmachineinterface |