Thing Translator: An Efficient Way to Classify Different Things

Abstract

Translation is the communication of a source-language text into an equivalent target-language text. A translator always risks in advertently introducing source-language words, grammar, or syntax into the target-language rendering. Thing Translator is an open-source offline application that uses the modern machine learning techniques – namely, computer vision and natural language translation. Thing Translator is a web app that lets you point your phone at particular object to hear it in a different language. The task of the translator here is to maintain a delicate balance between the source language and the language of translation. Thing translator is using Google’s Translate API's and cloud vision as it was developed by Google as a part of AI experiment. The main reason behind building this application is to help each and every person to know different things in many languages and understand them better. When we want to travel outside our place or outside India then there may be difficulty that we may not know that language of that particular region, we find language barrier due to not understanding the language they speak there is also a language barrier if want to shop from another region. We have designed an application ―Thing Translator: An Efficient Way to Classify Different Things‖ for every person to classify or recognize different things present around them. Thing translator is an android application which uses image recognition and object character recognition. This is useful to understand and to learn names of different objects just a capturing the image and also to learn pronunciation of the objects in native or other languages. This application can be used for the classification of necessary household objects present around us such as bottle, clock, facemask, wrist watch, comb etc. and also some electronic objects present in house that include laptop, television, charger etc. Also, we have included some flowers and animals.