Background: Over the past few decades, one of the most common issues has been visual impairments. People with visual impairments typically ask for assistance from others in order to continue performing their everyday responsibilities. They struggle to function in unpredictable or strange circumstances. These difficulties restrict their freedom and increase the disparity between them and others who are typically sighted.

Objectives: This research seeks to bridge the gap by producing a real-time speech-to-text pipeline tightly integrated with accurate digital text-to-Braille patterns and low-cost tactile rende

Methods: This study uses a three-phase methodology: first, speech recognition uses Google's Speech API to record Gujarati audio input and translate it into text. Based on the input speech, a text processing module then prepares and cleans the recognized Gujarati script. Lastly, each Gujarati character is mapped to its Bharati Braille Unicode equivalent via the Braille translation module, producing usable Braille output for persons with visual impairments.

Result Discussion : Using one, two, and three-word Gujarati sentences, the experimental evaluation was carried out independently for male and female speakers. The overall recognition accuracy was 88.57%. For brief Gujarati utterances, male participants obtained 100%, 83.33%, and 80% accuracy, whereas female participants achieved 100%, 83.33%, and 83.33%, respectively...