Howl: A Deployed, Open-Source Wake Word Detection System

Tang, Rachel; Lee, Jaejun; Razi, Afsaneh; Cambre, Julia; Bicking, Ian; Kaye, Jofish; Lin, Jimmy

doi:10.48550/arxiv.2008.09606

Cited by 1 publication

(2 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Most recently, Tang et al [19] have released Howl -a productionalized, open-source wake word detection toolkit, explored a number of models and achieved nearly-SOTA accuracy with a residual convolutional network architecture.…”

Section: A Previous Work On Kwsmentioning

confidence: 99%

See 1 more Smart Citation

Learning Efficient Representations for Keyword Spotting with Triplet Loss

Vygon

Mikhaylovskiy

2021

Lecture Notes in Computer Science

View full text Add to dashboard Cite

In the past few years, triplet loss-based metric embeddings have become a de-facto standard for several important computer vision problems, most notably, person reidentification. On the other hand, in the area of speech recognition the metric embeddings generated by the triplet loss are rarely used even for classification problems. We fill this gap showing that a combination of two representation learning techniques: a triplet loss-based embedding and a variant of kNN for classification instead of cross-entropy loss significantly (by 26% to 38%) improves the classification accuracy for convolutional networks on a LibriSpeech-derived LibriWords datasets. To do so, we propose a novel phonetic similarity based triplet mining approach. We also match the current best published SOTA for Google Speech Commands dataset V2 10+2-class classification with an architecture that is about 6 times more compact and improve the current best published SOTA for 35class classification on Google Speech Commands dataset V2 by over 40%. 1

show abstract

Section: A Previous Work On Kwsmentioning

confidence: 99%

“…We have mostly experimented with ResNet-based models res8 [19] and res15 [37]. The initial experiments have shown that RNN-based architectures show significantly worse results when trained for the triplet loss, so they were discarded in our later work.…”

Section: Model Architecturesmentioning

confidence: 99%

Learning Efficient Representations for Keyword Spotting with Triplet Loss

Vygon

Mikhaylovskiy

2021

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

Howl: A Deployed, Open-Source Wake Word Detection System

Cited by 1 publication

References 10 publications

Learning Efficient Representations for Keyword Spotting with Triplet Loss

Learning Efficient Representations for Keyword Spotting with Triplet Loss

Contact Info

Product

Resources

About