Analysis of Lip-Reading Using Deep Learning Techniques: A Review

Communication is a crucial aspect that occurs between a couple of people. Speech, consisting of audio-visual information, allows us to understand each other. Although individuals mainly rely on audio information to comprehend speech, visual details can enhance speech recognition, especially in noisy environments. Lip-reading is a phenomenon that can aid speech recognition systems by identifying the text spoken by an individual. To facilitate lip-reading, Automatic Lip-Reading (ALR) systems have been produced, using Artificial Intelligence (AI) techniques such as Machine Learning (ML) and Deep Learning (DL). This paper discusses the background behind lip-reading and delves into the workings of ALR systems, including the various DL and ML algorithms and popular English datasets used. The article presents a brief review of experiments conducted by other authors. Finally, the article is concluded by providing insights into the future scope of work.