Communication barriers between the hearing and deaf individuals that continue to limit accessibility and social inclusion in an education, healthcare, workplaces and also public environments. This research presents a Real-Time Speech-to-Glyph Translator that converts spoken language into visual glyphs instantly and correctly, enabling deaf and hard-of-hearing individuals to understand the verbal communication more effectively and accurately. The proposed system utilizes Automatic Speech Recognition (ASR) to capture and transcribe speech, followed by Natural Language Processing (NLP) techniques to analyze the recognized text and map it to an appropriate sequence of standardized glyphs. These translated glyphs are displayed through an intuitive graphical interface that ensure fast, accurate and User-friendly communication. This system is designed to minimize the translation latency while maintaining high recognition accuracy under multiple speaking conditions. provides real-time visual representation of spoken language and the proposed solution enhances accessibility and promotes inclusive communication and also reduces dependency on human interpreters. Experimental evaluation demonstrates that the system delivers reliable performance with low response time, making it suitable for educational institutions, public service centers, healthcare facilities and everyday communication. The research contributes to the assistive technology by offering an efficient and scalable solution for that bridges the communication gap between hearing and deaf communities and normal people.

The proposed Project “Real-Time Speech-to-Glyph Translator for Deaf Accessibility” system provides a effective assistive communication solution for the deaf and hard-of-hearing individuals by converting spoken language into visual glyph representations in real time with minimal delay. The system integrates Speech Recognition, Natural Language Processing (NLP) and graphical translation techniques to reduce the communication barriers and improve accessibility. The developed model enables faster and more interactive communication between the hearing and hearing-impaired individuals in various environments such as education institutions, workplaces, healthcare systems and public services. This proposed system also demonstrates the practical application of Artificial Intelligence and assistive technologies in promoting social inclusion and also equal communication opportunities. This project achieves improved communication efficiency, real-time performance and user accessibility. Future enhancements may include multilingual support, advanced gesture recognition and integration with wearable or mobile devices for improved portability and usability for those people.

1. This paper presents a web-based speech-to-text system that leverages the Web Speech API for in-browser real-time transcription, highlighting how client-side speech recognition interfaces can be implemented.https://www.ijfmr.com/papers/2023/2/1957.pdf 2. Benchmarking open source and paid services for speech to text – A formal evaluation of open source vs. commercial speech-to-text engines using metrics like Word Error Rate (WER), useful for comparing performance approaches for your system.,” https://www.frontiersin.org/journals/big-data/articles/10.3389/fdata.2023.1210559/full 3. A 2025 article describing automated pipelines for speech recognition and conversion into text, useful for discussing real-time pipelines and model evaluation for STT tasks.,” https://indjst.org/articles/systematic-approach-for-speech-to-text-translation-and- summarization 4. Automatic speech recognition and the transcription of … – A 2024 study com- paring ASR systems like Whisper and others on transcription accuracy, il- lustrating real-world strengths and limitations of contemporary STT tools. ,” https://www.frontiersin.org/journals/communication/articles/10.3389/fcomm.2024.1281407/full

© 2025 International Journal of Engineering and Techniques (IJET).

Digital Object Identifier (DOI)

IJET assigns a unique DOI to every accepted article via Zenodo for persistent linking and citation. Your article’s DOI: https://doi.org/{{doi}}

Open DOI About Zenodo DOI

Close