Implantation of a Reading Aid for Bengali-Speaking Individuals Who are Blind

Abstract

The reading aid is a hardware component with the necessary software installed to read printed materials aloud as a human reader would. Reading any printed material—books, documents, or anything else—is impossible for people with visual problems. Digitizing the reading materials, extracting the text from the image, and creating audio files from them will quickly solve this problem. This will allow people to understand the text on the paper merely by hearing the audio files created from the text. For those who speak Bengali and are visually impaired, a device has been implemented. The optical character identification (Oci) engine utilised for text detection is called tesser-act-oci. The Python-gtts module, a text-to-speech engine, is used to convert the words extracted by tesser-act-oci into sound.The entire operation is carried out using a small raspberry pi computer. Words to sound conversion and character detection accuracy from a captured image are both up to 85%. It should be remembered that the proportion of the right words to all the other words in a picture determines correctness.