# Quaternion Neural Networks for Multi-channel Distant Speech Recognition

arXiv: https://arxiv.org/abs/2005.08566 · HTML: https://mlsanity.com/paper/2005.08566/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 18 May 2020.

## Built on by (most cited first, 5 of 5)

- [Convolutional Filtering and Neural Networks with Non Commutative Algebras](https://mlsanity.com/paper/2108.09923/) (2021)
- [Learning Speech Emotion Representations in the Quaternion Domain](https://mlsanity.com/paper/2204.02385/) (2022)
- [BERSting at the Screams: A Benchmark for Distanced, Emotional and Shouted Speech Recognition](https://mlsanity.com/paper/2505.00059/) (2025)
- [Exploiting Single-Channel Speech for Multi-Channel End-to-End Speech Recognition: A Comparative Study](https://mlsanity.com/paper/2203.16757/) (2022)
- [QC-GAN: A Parameter-Efficient Quaternion Conformer GAN for High-Fidelity Speech Enhancement](https://mlsanity.com/paper/2606.18611/) (2026)

## Abstract (arXiv)

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.