AutoKWS: Keyword Spotting with Differentiable Architecture Search

by   Bo Zhang, et al.

Smart audio devices are gated by an always-on lightweight keyword spotting program to reduce power consumption. It is however challenging to design models that have both high accuracy and low latency for accurate and fast responsiveness. Many efforts have been made to develop end-to-end neural networks, in which depthwise separable convolutions, temporal convolutions, and LSTMs are adopted as building units. Nonetheless, these networks designed with human expertise may not achieve an optimal trade-off in an expansive search space. In this paper, we propose to leverage recent advances in differentiable neural architecture search to discover more efficient networks. Our found model attains 97.2


Neural Architecture Search For Keyword Spotting

Deep neural networks have recently become a popular solution to keyword ...

Small-Footprint Keyword Spotting on Raw Audio Data with Sinc-Convolutions

Keyword Spotting (KWS) enables speech-based user interaction on smart de...

Performance-Oriented Neural Architecture Search

Hardware-Software Co-Design is a highly successful strategy for improvin...

Encoder-Decoder Neural Architecture Optimization for Keyword Spotting

Keyword spotting aims to identify specific keyword audio utterances. In ...

Stochastic Adaptive Neural Architecture Search for Keyword Spotting

The problem of keyword spotting i.e. identifying keywords in a real-time...

Latency-Aware Differentiable Neural Architecture Search

Differentiable neural architecture search methods became popular in auto...

FuSeConv: Fully Separable Convolutions for Fast Inference on Systolic Arrays

Both efficient neural networks and hardware accelerators are being explo...