LCANet: End-to-End Lipreading with Cascaded Attention-CTC

TL;DR

LCANet is an end-to-end lipreading system that combines CNN, highway, and bidirectional GRU encoders with a cascaded attention-CTC decoder, significantly improving accuracy on benchmark datasets.

Contribution

The paper introduces LCANet, a novel deep neural network architecture with a cascaded attention-CTC decoder that enhances lipreading performance and convergence speed.

Findings

Achieves 1.3% CER and 3.0% WER on GRID corpus

Improves state-of-the-art results by 12.3%

Effectively captures spatio-temporal information in videos

Abstract

Machine lipreading is a special type of automatic speech recognition (ASR) which transcribes human speech by visually interpreting the movement of related face regions including lips, face, and tongue. Recently, deep neural network based lipreading methods show great potential and have exceeded the accuracy of experienced human lipreaders in some benchmark datasets. However, lipreading is still far from being solved, and existing methods tend to have high error rates on the wild data. In this paper, we propose LCANet, an end-to-end deep neural network based lipreading system. LCANet encodes input video frames using a stacked 3D convolutional neural network (CNN), highway network and bidirectional GRU network. The encoder effectively captures both short-term and long-term spatio-temporal information. More importantly, LCANet incorporates a cascaded attention-CTC decoder to generate output texts. By cascading CTC with attention, it partially eliminates the defect of the conditional independence assumption of CTC within the hidden neural layers, and this yields notably performance improvement as well as faster convergence. The experimental results show the proposed system achieves a 1.3% CER and 3.0% WER on the GRID corpus database, leading to a 12.3% improvement compared to the state-of-the-art methods.