Audio Lens–Assistive Smart Glass for Blind

Audio Lens is an assistive technology device to increase independence and access to printed materials and object recognition for those who have an inoperable vision impairment. Audio Lens is a Raspberry Pi platform that uses modern scanning and AI technologies to allow independence to be increased for the user. The user captures an image of printed material from books, menus, signage, etc., using a high-resolution camera, and sends the image to Tesseract OCR which processes the image into text that can be output audibly using a Text-to-Speech (TTS) program, so the user can produce content for consumption without the need of an able-bodied person to read to the individual. The device also has a real-time object detection process that utilizes the YOLO-model to allow someone to understand objects that exist in their world, providing situational awareness and safer navigation. The device audibly announces objects detected, subsequently providing another source of inputs useful for interaction in the environment. The device utilizes a rechargeable battery that allows portability and convenience for daily use. The device contains very simple controls that a user can access through basic button presses or detection gestures, allowing the technical and developmental controls to be simple enough to allow anyone to use at any age or technical level. Audio Lens developed a useful, wearable solution that empowers mobility, safety, and inclusion for people who are visually impaired by combining text-to-speech and object recognition technology, allowing them to close the accessibility gap and move confidently through the world.