Broadcasted Residual Learning for Efficient Keyword Spotting

by   Byeonggeun Kim, et al.

Keyword spotting is an important research field because it plays a key role in device wake-up and user interaction on smart devices. However, it is challenging to minimize errors while operating efficiently in devices with limited resources such as mobile phones. We present a broadcasted residual learning method to achieve high accuracy with small model size and computational load. Our method configures most of the residual functions as 1D temporal convolution while still allows 2D convolution together using a broadcasted-residual connection that expands temporal output to frequency-temporal dimension. This residual mapping enables the network to effectively represent useful audio features with much less computation than conventional convolutional neural networks. We also propose a novel network architecture, Broadcasting-residual network (BC-ResNet), based on broadcasted residual learning and describe how to scale up the model according to the target device's resources. BC-ResNets achieve state-of-the-art 98.0 top-1 accuracy on Google speech command datasets v1 and v2, respectively, and consistently outperform previous approaches, using fewer computations and parameters.



There are no comments yet.


page 1

page 2

page 3

page 4


Deep Residual Learning for Small-Footprint Keyword Spotting

We explore the application of deep residual learning and dilated convolu...

Temporal Convolution for Real-time Keyword Spotting on Mobile Devices

Keyword spotting (KWS) plays a critical role in enabling speech-based us...

URNet : User-Resizable Residual Networks with Conditional Gating Module

Convolutional Neural Networks are widely used to process spatial scenes,...

Domain Generalization on Efficient Acoustic Scene Classification using Residual Normalization

It is a practical research topic how to deal with multi-device audio inp...

Keyword Spotting for Hearing Assistive Devices Robust to External Speakers

Keyword spotting (KWS) is experiencing an upswing due to the pervasivene...

Small-Footprint Keyword Spotting with Multi-Scale Temporal Convolution

Keyword Spotting (KWS) plays a vital role in human-computer interaction ...

Exploring Filterbank Learning for Keyword Spotting

Despite their great performance over the years, handcrafted speech featu...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.