Does Single-channel Speech Enhancement Improve Keyword Spotting Accuracy? A Case Study

no code implementations • 27 Sep 2023 • Avamarie Brueggeman, Takuya Higuchi, Masood Delfarah, Stephen Shum, Vineet Garg

Our investigation reveals that SE can improve KWS accuracy on noisy speech when the backend model is trained on clean speech; however, despite our extensive exploration, it is difficult to improve the KWS accuracy with SE when the backend is trained on noisy speech.

Automatic Speech Recognition Keyword Spotting +3

Multichannel Voice Trigger Detection Based on Transform-average-concatenate

no code implementations • 27 Sep 2023 • Takuya Higuchi, Avamarie Brueggeman, Masood Delfarah, Stephen Shum

Voice triggering (VT) enables users to activate their devices by just speaking a trigger phrase.

Speech Enhancement

Improving Voice Trigger Detection with Metric Learning

no code implementations • 5 Apr 2022 • Prateeth Nayak, Takuya Higuchi, Anmol Gupta, Shivesh Ranjan, Stephen Shum, Siddharth Sigtia, Erik Marchi, Varun Lakshminarasimhan, Minsik Cho, Saurabh Adya, Chandra Dhir, Ahmed Tewfik

A detector is typically trained on speech data independent of speaker information and used for the voice trigger detection task.

Metric Learning

Improving on-device speaker verification using federated learning with privacy

no code implementations • 6 Aug 2020 • Filip Granqvist, Matt Seigel, Rogier Van Dalen, Áine Cahill, Stephen Shum, Matthias Paulik

From these features, the model predicts speaker characteristic labels considered useful as side information.

Federated Learning Multi-Task Learning +3