Predicting detection filters for small footprint open-vocabulary keyword spotting

TL;DR

This paper introduces a neural network-based approach for open-vocabulary keyword spotting that is lightweight, customizable, and does not require task-specific data, outperforming traditional methods in various detection tasks.

Contribution

A novel neural network model predicts detection filters for any keyword, enabling customizable voice interfaces without task-specific training data.

Findings

Model outperforms acoustic keyword spotting baselines

Supports fine-tuning for specific keywords

Maintains performance on new keywords after fine-tuning

Abstract

In this paper, we propose a fully-neural approach to open-vocabulary keyword spotting, that allows the users to include a customizable voice interface to their device and that does not require task-specific data. We present a keyword detection neural network weighing less than 250KB, in which the topmost layer performing keyword detection is predicted by an auxiliary network, that may be run offline to generate a detector for any keyword. We show that the proposed model outperforms acoustic keyword spotting baselines by a large margin on two tasks of detecting keywords in utterances and three tasks of detecting isolated speech commands. We also propose a method to fine-tune the model when specific training data is available for some keywords, which yields a performance similar to a standard speech command neural network while keeping the ability of the model to be applied to new keywords.