In this work, we propose Tiny-CRNN (Tiny Convolutional Recurrent Neural Network) models applied to the problem of wakeword detection, and augment them with scaled dot product attention. We find that, compared to Convolutional Neural Network models, False Accepts in a 250k parameter budget can be reduced by 25% with a 10% reduction in parameter size by using models based on the Tiny-CRNN architecture, and we can get up to 32% reduction in False Accepts at a 50k parameter budget with 75% reduction in parameter size compared to word-level Dense Neural Network models. We discuss solutions to the challenging problem of performing inference on streaming audio with this architecture, as well as differences in start-end index errors and latency in comparison to CNN, DNN, and DNN-HMM models.
@article{arxiv.2109.14725,
title = {Tiny-CRNN: Streaming Wakeword Detection In A Low Footprint Setting},
author = {Mohammad Omar Khursheed and Christin Jose and Rajath Kumar and Gengshen Fu and Brian Kulis and Santosh Kumar Cheekatmalla},
journal= {arXiv preprint arXiv:2109.14725},
year = {2021}
}
Comments
arXiv admin note: substantial text overlap with arXiv:2011.12941