Skip to content
CaMLSysUniversity of Cambridge
Home People Publications Projects Contact

← All publications

Quaternion Neural Networks for Multi-Channel Distant Speech Recognition

  • Xinchi QiuUniversity of Oxford
  • Titouan ParcolletUniversity of Oxford
  • Mirco RavanelliMilaUniversité de Montréal
  • Nicholas D. LaneUniversity of OxfordSamsung AI, Cambridge
  • Mohamed MorchidAvignon University

Interspeech, 2020

Abstract

Despite the significant progress in automatic speech recognition (ASR), distant ASR remains challenging due to noise and reverberation. A common approach to mitigate this issue consists of equipping the recording devices with multiple microphones that capture the acoustic scene from different perspectives. These multi-channel audio recordings contain specific internal relations between each signal. In this paper, we propose to capture these inter- and intra- structural dependencies with quaternion neural networks, which can jointly process multiple signals as whole quaternion entities. The quaternion algebra replaces the standard dot product with the Hamilton one, thus offering a simple and elegant way to model dependencies between elements. The quaternion layers are then coupled with a recurrent neural network, which can learn long-term dependencies in the time domain. We show that a quaternion long-short term memory neural network (QLSTM), trained on the concatenated multi-channel speech signals, outperforms equivalent real-valued LSTM on two different tasks of multi-channel distant speech recognition.

Materials

  • PDF
BibTeX
@inproceedings{qiu2020quaternion,
  title = {{Quaternion Neural Networks for Multi-Channel Distant Speech Recognition}},
  author = {Qiu, Xinchi and Parcollet, Titouan and Ravanelli, Mirco and Lane, Nicholas D. and Morchid, Mohamed},
  year = {2020},
  booktitle = {Interspeech},
  pages = {329--333},
  publisher = {ISCA},
  doi = {10.21437/interspeech.2020-1682},
  url = {https://doi.org/10.21437/interspeech.2020-1682}
}
Cambridge Machine Learning Systems Group
Department of Computer Science and Technology
University of Cambridge
ContactGitHub