Deep Residual Learning for Small-Footprint Keyword Spotting
TL
TL;DR
This work explores the application of deep residual learning and dilated convolutions to the keyword spotting task, using the recently-released Google Speech Commands Dataset as a benchmark and establishes an open-source state-of-the-art reference to support the development of future speech-based interfaces.
Abstract
We explore the application of deep residual learning and dilated convolutions to the keyword spotting task, using the recently-released Google Speech Commands Dataset as our benchmark. Our best residual network (ResNet) implementation significantly outperforms Google's previous convolutional neural networks in terms of accuracy. By varying model depth and width, we can achieve compact models that also outperform previous small-footprint variants. To our knowledge, we are the first to examine these approaches for keyword spotting, and our results establish an open-source state-of-the-art reference to support the development of future speech-based interfaces.
References12 items
1
Honk: A PyTorch Reimplementation of Convolutional Neural Networks for Keyword Spotting
2
The Microsoft 2017 Conversational Speech Recognition System
3
End-to-End Text-Independent Speaker Verification with Triplet Loss on Short Utterances
4
Convolutional Recurrent Neural Networks for Small-Footprint Keyword Spotting
5