Lip Movement Detection—A Survey of State-of-the-Art Approaches

New technologies such as machine learning (ML) and artificial Intelligence (AI) are gaining more attention to make several tasks easier and efficient, one of which is converting speech to text and vice versa. Much research has been conducted in the field of deep learning and natural language processing to gain increased efficiency in detecting words and phrases via detecting lip movements. The reason behind this increasing interest in lip-reading techniques is its versatile real-world applications like accessibility to people with hearing and speaking disability, efficiency in voice search, new ways to add biometric security, human–computer interaction, education, etc. This paper aims to highlight some of the key features of different research conducted, comparing them, getting to know the key milestones set throughout the past decades and identifying emerging trends in this field which will help to suggest the potential future direction in the same. Some of the key features discussed in this paper are different algorithms and methodologies adopted, the dataset used, performance matrices, and their pros and cons in different real-world applications.