Deep neural network based wake-up-word speech recognition with two-stage detection

Fengpei Ge, Yonghong YanICASSP 2017 · OpenReview

Abstract

This paper presents a novel far-field voice trigger algorithm utilizing DNN with the objective function of state-level minimum Bayes risk for training, customizing the decoding network to absorb the ambient noise and background speech. We adopt a two-stage classification strategy to integrate the phonetic knowledge and model-based classification into detecting wake-up words. Experimental results of the online test show that it can provide a higher than 90% accuracy and meanwhile false alarms are less than once per nine hours in the noisy home environments where the sound pressure level is about 80dB.
BibTeX
@inproceedings{ge2017deep,
  author = {Fengpei Ge and Yonghong Yan},
  title = {{Deep neural network based wake-up-word speech recognition with two-stage detection}},
  booktitle = {IEEE International Conference on Acoustics, Speech and Signal Processing},
  year = {2017},
  url = {https://openreview.net/forum?id=U16T8qMyvp},
}

Loading related content...

Figure