Explore the N-Gram Model of Adaptive Prediction Text

Abstract

This program is aimed towards enabling people with speaking disabilities to participate more within their conversations. People with speaking disabilities are often forced to rely upon sign language or some form of text to speech to communicate in their day-to-day life. This often creates trouble as some people may not know sign language, and typing out every single thing you may want to say takes a lot of time and effort. This program will help these issues by creating a text-to-speech text generator. By using pattern recognition, the program will learn the person’s talking styles, and be able to more fluently autofill the sentence; thereby requiring less effort and time on the user. The program uses a probabilistic n-gram model in order to predict what the user might want to say in real time. By using the user’s input as training data in the future, the n-gram models can adapt to the style and tone of the user reasonably quickly.