TTS-Assistive Technology in Traffic Road Sign Recognition (TRSR) for Visually Impaired

Abstract

Visual Impairment has always impeded an individual in the normal functioning of their daytoday activities. Their woes and troubles are immeasurable, especially in public places. Reading road signs to ensure their safety while commuting on roads and crossings has always been challenging. Even though Braille and other assistive technologies are very much prevalent to help them in becoming independent, the increase in road accidents every day makes road safety an ordeal for visually challenged pedestrians. According to a 2021 report by the International Agency for the Prevention of Blindness's Vision Atlas, there are 43 million people with onehundred percent blindness and 295 million people living with moderate-to-severe visual impairment. Text-To-Speech (TTS) technology is an assistive technology that converts text into speech. The standard of a speech synthesizer is judged by its similarity to the human voice and by its ability to be understood clearly. TTS is used to create an audio output within the type of a spoken voice. Text is complementary to different visual cues, like contour, color, and texture. Text Detection is a task to detect text regions in a complex background and label them with bounding boxes, in natural scenes. Text recognition is the strategy of converting text regions into computer-readable forms. The challenges of text recognition are the diversity of scene text, the complexity of background, and various interference factors. In this work, TTS technology is suggested, which aids the visually challenged in maneuvering roads with ease and confidence. The efficacy of text recognition is compared based on parameters such as Precision, Recall, F1 score, and Accuracy.