Optical Text Translator from Images using Machine Learning

Abstract

In the actual world, when people go to a new location, they may have difficulties with language, signboards, or other kinds of communication. People in the digital age utilize Google Translate, Pocket Dictionary, and other similar applications. A signboard or other kind of notice may communicate important information or even act as a warning. Important information may be lost if the message is unavailable to individuals of different linguistic backgrounds; furthermore, visitors may struggle to carry out their duties in a new place if they do not comprehend the language spoken there. A portable dictionary, on the other hand, may be ineffective if users wish to translate a language in which words are not organized alphabetically. In another research, it is also understood in the same manner that users are unable to type the text of what they see. Because they are unable to comprehend the language despite the pocket dictionary and online translation service offered, people from various linguistic backgrounds may experience a breakdown in communication. The purpose of this article is to create an app that will translate information shown on a signboard into the language of the user's choice. When the user points their phone's camera at the signboard, the App takes a picture of it and then translates the text on the signboard.