HANDWRITING TEXT RECOGNITION USING NEURAL NETWORK

2022 International Research Journal of Modernization in Engineering Technology and Science  
This project introduces a way to recognize text or handwritten information covert into machine readable text in a form the computer can process, store and edit as a text file or as a part of a data entry and manipulation software. Then it converts the machine readable text to speech. People who suffer from low vision, sight and visual impairments, who are not able to see words and letters in ordinary newsprint, books and notes were the ones this proposal was meant to help. Moreover this project
more » ... also can be capstone for research and development of artificially intelligent virtual assistants that require a glimpse into the human world for information such as street signs and receipts. Using a combination of Convolution neural networks and Recurrent Neural Networks the project successfully came about to recognize text and hand written information. Another deep learning network, WaveNet was employed to convert the text which is recognized from the handwriting to covert into a raw audio file. Hence using this complete system one can pass an image containing text, have it recognized by the computer and also be read out to the user.
doi:10.56726/irjmets29660 fatcat:qf3vsnkrsnbc3mydkrbadhn5d4