Open AccessText Recognition Using Silent Speech
Author(s) -
Girish Kumar
Publication year - 2021
Publication title -
international journal for research in applied science and engineering technology
Language(s) - English
Resource type - Journals
ISSN - 2321-9653
DOI - 10.22214/ijraset.2021.37414
Subject(s) - computer science , focus (optics) , speech recognition , pipeline (software) , set (abstract data type) , convolution (computer science) , convolutional neural network , artificial intelligence , natural language processing , data set , motion (physics) , artificial neural network , physics , optics , programming language
Our objective is to identify the characters from the quite speech of the English language. We tend to focus on the lip region to recognize the characters spoken clearly in the video. Our contribution is: foremost, this model is developed by using a pipeline method form absolutely automatic information assortment from the video. Though this, it generates a data set that is spoken by the individuals. Secondly, it is developed by using the machine learning algorithm Convolution Neural Network (CNN) that learns the lip motion. Thirdly, Convolution network turn out the efficient result by examining the video and also the data set.