# Data Augmentation for Robust Keyword Spotting under Playback Interference

arXiv: https://arxiv.org/abs/1808.00563 · HTML: https://mlsanity.com/paper/1808.00563/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 1 Aug 2018.

## Built on by (most cited first, 8 of 8)

- [Gesture Recognition for FMCW Radar on the Edge](https://mlsanity.com/paper/2310.08876/) (2023)
- [Text Anchor Based Metric Learning for Small-footprint Keyword Spotting](https://mlsanity.com/paper/2108.05516/) (2021)
- [Spot keywords from very noisy and mixed speech](https://mlsanity.com/paper/2305.17706/) (2023)
- [Robust Wake-Up Word Detection by Two-stage Multi-resolution Ensembles](https://mlsanity.com/paper/2310.11379/) (2023)
- [Behavior of Keyword Spotting Networks Under Noisy Conditions](https://mlsanity.com/paper/2109.07930/) (2021)
- [AB/BA analysis: A framework for estimating keyword spotting recall improvement while maintaining audio privacy](https://mlsanity.com/paper/2204.08474/) (2022)
- [AraSpot: Arabic Spoken Command Spotting](https://mlsanity.com/paper/2303.16621/) (2023)
- [GraphemeAug: A Systematic Approach to Synthesized Hard Negative Keyword Spotting Examples](https://mlsanity.com/paper/2505.14814/) (2025)

## Abstract (arXiv)

Accurate on-device keyword spotting (KWS) with low false accept and false reject rate is crucial to customer experience for far-field voice control of conversational agents. It is particularly challenging to maintain low false reject rate in real world conditions where there is (a) ambient noise from external sources such as TV, household appliances, or other speech that is not directed at the device (b) imperfect cancellation of the audio playback from the device, resulting in residual echo, after being processed by the Acoustic Echo Cancellation (AEC) system. In this paper, we propose a data augmentation strategy to improve keyword spotting performance under these challenging conditions. The training set audio is artificially corrupted by mixing in music and TV/movie audio, at different signal to interference ratios. Our results show that we get around 30-45% relative reduction in false reject rates, at a range of false alarm rates, under audio playback from such devices.