DocuVoice: Survey of Intelligent Document Analysis Solutions for the Visually Impaired

The availability of written information is a challenge for visually impaired people. The traditional screen reader and text-to-speech technology enable linear reading but do not sup- port understanding and interaction. This paper presents a survey of the current state of the art and research in document anal- ysis and accessibility, including Optical Character Recognition, layout analysis, Natural Language Processing, summarization, and speech interfaces. The survey highlights the key gaps in the current state of the art, including dealing with complex layouts, explaining figures and tables, multilingual support, and data privacy. This paper also introduces a unified framework for an intelligent document analyzer using AI technology called Docu- Voice.