论文信息 - Neural ODE with Temporal Convolution and Time Delay Neural Networks for Small-Footprint Keyword Spotting

Neural ODE with Temporal Convolution and Time Delay Neural Networks for Small-Footprint Keyword Spotting

In this paper, we propose neural network models based on the neural ordinary differential equation (NODE) for small-footprint keyword spotting (KWS). We present techniques to apply NODE to KWS that make it possible to adopt Batch Normalization to NODE-based network and to reduce the number of computations during inference. Finally, we show that the number of model parameters of the proposed model is smaller by 68% than that of the conventional KWS model.

Hiroshi Fuketa | Yukinori Morita

[1] Sergey Ioffe,et al. Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

[2] Yee Whye Teh,et al. Augmented Neural ODEs , 2019, NeurIPS.

[3] Jimmy J. Lin,et al. Deep Residual Learning for Small-Footprint Keyword Spotting , 2017, 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[4] David Duvenaud,et al. Neural Ordinary Differential Equations , 2018, NeurIPS.

[5] Dongyoung Kim,et al. Temporal Convolution for Real-time Keyword Spotting on Mobile Devices , 2019, INTERSPEECH.

[6] Yundong Zhang,et al. Hello Edge: Keyword Spotting on Microcontrollers , 2017, ArXiv.

[7] Sanjeev Khudanpur,et al. A time delay neural network architecture for efficient modeling of long temporal contexts , 2015, INTERSPEECH.

[8] Tara N. Sainath,et al. Convolutional neural networks for small-footprint keyword spotting , 2015, INTERSPEECH.

[9] Jiangyan Yi,et al. A Time Delay Neural Network with Shared Weight Self-Attention for Small-Footprint Keyword Spotting , 2019, INTERSPEECH.