Quaternion Convolutional Neural Networks for End-to-End Automatic Speech Recognition

TL;DR

This paper introduces quaternion-valued convolutional neural networks for end-to-end speech recognition, leveraging quaternion algebra to process multidimensional features more effectively than traditional real-valued models.

Contribution

It proposes integrating multiple feature views into quaternion CNNs for sequence-to-sequence speech recognition with CTC, demonstrating improved performance with fewer parameters.

Findings

Lower phoneme error rate (PER) on TIMIT corpus

Fewer learning parameters needed compared to real-valued CNNs

Effective processing of multidimensional speech features

Abstract

Recently, the connectionist temporal classification (CTC) model coupled with recurrent (RNN) or convolutional neural networks (CNN), made it easier to train speech recognition systems in an end-to-end fashion. However in real-valued models, time frame components such as mel-filter-bank energies and the cepstral coefficients obtained from them, together with their first and second order derivatives, are processed as individual elements, while a natural alternative is to process such components as composed entities. We propose to group such elements in the form of quaternions and to process these quaternions using the established quaternion algebra. Quaternion numbers and quaternion neural networks have shown their efficiency to process multidimensional inputs as entities, to encode internal dependencies, and to solve many tasks with less learning parameters than real-valued models. This paper proposes to integrate multiple feature views in quaternion-valued convolutional neural network (QCNN), to be used for sequence-to-sequence mapping with the CTC model. Promising results are reported using simple QCNNs in phoneme recognition experiments with the TIMIT corpus. More precisely, QCNNs obtain a lower phoneme error rate (PER) with less learning parameters than a competing model based on real-valued CNNs.