by Nimish Jain
Communication is one of the most essential prospects of life and yet over 18 million Americans and 500,000 Canadians with speech impairments lack access to an effective communication method. Current solutions developed to address this issue are slow, rendering them commercially unviable. This work is mainly motivated to give these people a voice by providing a free, accessible and easy-to-use solution. By employing a variety of novel machine learning techniques, HELLO analyzes the lip movements in a given soundless video and converts it into text and a playable audio file.