# Tiny-CRNN: Streaming Wakeword Detection In A Low Footprint Setting

arXiv: https://arxiv.org/abs/2109.14725 · HTML: https://mlsanity.com/paper/2109.14725/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 6 of its 15 references that are in the map.
- First posted on arXiv 29 Sep 2021.

## Builds on (most cited first, 6 of 6)

- [Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling](https://mlsanity.com/paper/1412.3555/) (2014)
- [Streaming keyword spotting on mobile devices](https://mlsanity.com/paper/2005.06720/) (2020)
- [Convolutional Recurrent Neural Networks for Small-Footprint Keyword Spotting](https://mlsanity.com/paper/1703.05390/) (2017)
- [Accurate Detection of Wake Word Start and End Using a CNN](https://mlsanity.com/paper/2008.03790/) (2020)
- [Streaming Small-Footprint Keyword Spotting using Sequence-to-Sequence Models](https://mlsanity.com/paper/1710.09617/) (2017)
- [On Front-end Gain Invariant Modeling for Wake Word Spotting](https://mlsanity.com/paper/2010.06676/) (2020)

## Built on by (most cited first, 5 of 5)

- [A Fast Network Exploration Strategy to Profile Low Energy Consumption for Keyword Spotting](https://mlsanity.com/paper/2202.02361/) (2022)
- [Improving Voice Trigger Detection with Metric Learning](https://mlsanity.com/paper/2204.02455/) (2022)
- [Latency Control for Keyword Spotting](https://mlsanity.com/paper/2206.07261/) (2022)
- [Small-footprint slimmable networks for keyword spotting](https://mlsanity.com/paper/2304.12183/) (2023)
- [Self-supervised speech representation learning for keyword-spotting with light-weight transformers](https://mlsanity.com/paper/2303.04255/) (2023)

## Abstract (arXiv)

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.