Query-By-Example Keyword Spotting System Using Multi-Head Attention and Soft-triple Loss
Jinmiao Huang, Waseem Gharbieh, Han Suk Shim et al. · IEEE International Conference on Acoustics Speech and Signal Processing · 2021
This paper proposes a neural network architecture for tackling the query-by-example user-defined keyword spotting task. A multi-head attention module is added on top of a multi-layered GRU for effective feature extraction, and a normalized multi-head attention module is proposed for feature aggregation. We also adopt the softtriple loss - a combination of triplet loss and softmax loss - and showcase its effectiveness. We demonstrate the performance of our model on internal datasets with different languages and the public Hey-Snips dataset. We compare the performance of our model to a baseline system [1] and conduct an ablation study to show the benefit of each component in our architecture. The proposed work shows solid performance while preserving simplicity.
Cited by 35 works. doi:10.1109/icassp39728.2021.9414156
Authors
References
- Exploiting Generative AI to Scale up Intelligent Tutoring Systems
- Attention Is All You Need
- Librispeech: An ASR corpus based on public domain audio books
- In Defense of the Triplet Loss for Person Re-Identification
- Small-footprint keyword spotting using deep neural networks
- Montreal Forced Aligner: Trainable Text-Speech Alignment Using Kaldi
- Rapid and accurate spoken term detection
- Unsupervised spoken keyword spotting via segmental DTW on Gaussian posteriorgrams
- Query-by-example spoken term detection using phonetic posteriorgram templates
- SoftTriple Loss: Deep Metric Learning Without Triplet Sampling