Combining Multiple Views for Visual Speech Recognition

TL;DR

This paper investigates combining multiple camera views to enhance visual speech recognition, demonstrating significant performance improvements through various fusion strategies and view combinations.

Contribution

It provides a comprehensive analysis of multi-view fusion at feature and decision levels for visual speech recognition, highlighting the benefits of combining different camera angles.

Findings

Multi-view fusion improves recognition accuracy.

Combining views increases sentence correctness from 76% to 83%.

Fusion at decision level yields significant performance gains.

Abstract

Visual speech recognition is a challenging research problem with a particular practical application of aiding audio speech recognition in noisy scenarios. Multiple camera setups can be beneficial for the visual speech recognition systems in terms of improved performance and robustness. In this paper, we explore this aspect and provide a comprehensive study on combining multiple views for visual speech recognition. The thorough analysis covers fusion of all possible view angle combinations both at feature level and decision level. The employed visual speech recognition system in this study extracts features through a PCA-based convolutional neural network, followed by an LSTM network. Finally, these features are processed in a tandem system, being fed into a GMM-HMM scheme. The decision fusion acts after this point by combining the Viterbi path log-likelihoods. The results show that the complementary information contained in recordings from different view angles improves the results significantly. For example, the sentence correctness on the test set is increased from 76% for the highest performing single view ($30^\circ$) to up to 83% when combining this view with the frontal and $60^\circ$ view angles.