# Improving Voice Trigger Detection with Metric Learning

arXiv: https://arxiv.org/abs/2204.02455 · HTML: https://mlsanity.com/paper/2204.02455/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 36 references that are in the map.
- First posted on arXiv 5 Apr 2022.

## Builds on (most cited first, 19 of 19)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Deep Residual Learning for Small-Footprint Keyword Spotting](https://mlsanity.com/paper/1710.10361/) (2017)
- [Keyword Transformer: A Self-Attention Model for Keyword Spotting](https://mlsanity.com/paper/2104.00769/) (2021)
- [Temporal Convolution for Real-time Keyword Spotting on Mobile Devices](https://mlsanity.com/paper/1904.03814/) (2019)
- [Personalizing ASR for Dysarthric and Accented Speech with Limited Data](https://mlsanity.com/paper/1907.13511/) (2019)
- [MatchboxNet: 1D Time-Channel Separable Convolutional Neural Network Architecture for Speech Commands Recognition](https://mlsanity.com/paper/2004.08531/) (2020)
- [Convolutional Recurrent Neural Networks for Small-Footprint Keyword Spotting](https://mlsanity.com/paper/1703.05390/) (2017)
- [Query-by-Example Keyword Spotting system using Multi-head Attention and Softtriple Loss](https://mlsanity.com/paper/2102.07061/) (2021)
- [Streaming Small-Footprint Keyword Spotting using Sequence-to-Sequence Models](https://mlsanity.com/paper/1710.09617/) (2017)
- [Query-by-example on-device keyword spotting](https://mlsanity.com/paper/1910.05171/) (2019)
- [Max-Pooling Loss Training of Long Short-Term Memory Networks for Small-Footprint Keyword Spotting](https://mlsanity.com/paper/1705.02411/) (2017)
- [Multi-task Learning for Speaker Verification and Voice Trigger Detection](https://mlsanity.com/paper/2001.10816/) (2020)
- [Hybrid Transformer/CTC Networks for Hardware Efficient Voice Triggering](https://mlsanity.com/paper/2008.02323/) (2020)
- [Personalized Keyphrase Detection using Speaker and Environment Information](https://mlsanity.com/paper/2104.13970/) (2021)
- [Multi-task Learning for Voice Trigger Detection](https://mlsanity.com/paper/2001.09519/) (2020)
- [Tiny-CRNN: Streaming Wakeword Detection In A Low Footprint Setting](https://mlsanity.com/paper/2109.14725/) (2021)
- [The 2020 Personalized Voice Trigger Challenge: Open Database, Evaluation Metrics and the Baseline Systems](https://mlsanity.com/paper/2101.01935/) (2021)
- [The NPU System for the 2020 Personalized Voice Trigger Challenge](https://mlsanity.com/paper/2102.13552/) (2021)
- [Multi-task Learning with Cross Attention for Keyword Spotting](https://mlsanity.com/paper/2107.07634/) (2021)

## Built on by (most cited first, 1 of 1)

- [Efficient Multimodal Neural Networks for Trigger-less Voice Assistants](https://mlsanity.com/paper/2305.12063/) (2023)

## Abstract (arXiv)

Voice trigger detection is an important task, which enables activating a voice assistant when a target user speaks a keyword phrase. A detector is typically trained on speech data independent of speaker information and used for the voice trigger detection task. However, such a speaker independent voice trigger detector typically suffers from performance degradation on speech from underrepresented groups, such as accented speakers. In this work, we propose a novel voice trigger detector that can use a small number of utterances from a target speaker to improve detection accuracy. Our proposed model employs an encoder-decoder architecture. While the encoder performs speaker independent voice trigger detection, similar to the conventional detector, the decoder predicts a personalized embedding for each utterance. A personalized voice trigger score is then obtained as a similarity score between the embeddings of enrollment utterances and a test utterance. The personalized embedding allows adapting to target speaker's speech when computing the voice trigger score, hence improving voice trigger detection accuracy. Experimental results show that the proposed approach achieves a 38% relative reduction in a false rejection rate (FRR) compared to a baseline speaker independent voice trigger model.