Speech-impaired individuals face barriers in communication across education, healthcare & workplaces

Authors

  • Dr Rizwan Uz Zama Assistant Professor, Department of CSE, Lords Institute of Engineering and Technology, Hyderabad, India. Author
  • Ms Sameeha Firdaus, Ayesha Firdous, Khadeeja Soha Khan, Sumayya Fatima, Vijay Laxmi Students, Department of CSE, Lords Institute of Engineering and Technology, Hyderabad, India. Author

Keywords:

Sign language recognition; Assistive technology; Speech impairment; MediaPipe; Hand landmarks; BiLSTM; Attention; Text-to-speech; Speech-to-text; Human–computer interaction.

Abstract

Speech-impaired individuals face significant communication barriers in education, healthcare and workplace 
settings, where interaction increasingly depends on spoken language and trained sign-language interpreters are 
not always available. Existing assistive solutions are often hardware-dependent, expensive or limited to a fixed 
vocabulary, which makes them impractical for everyday use in resource-constrained institutions. This paper 
presents SignBridge, a lightweight, camera-based assistive communication system that recognises hand gestures 
and signs from a standard webcam and converts them into readable text and synthesised speech in real time, while 
converting spoken or typed replies into text for two-way communication. The system extracts hand and upper
body landmarks with MediaPipe Holistic, normalises them for position and scale, smooths them with a One-Euro 
filter and enriches them with joint-angle and velocity features. A hybrid one-dimensional CNN, bidirectional 
LSTM and temporal-attention network classifies sliding windows of landmarks, and a sign-spotting and temporal
voting stage turns the frame stream into stable sign decisions. Recognised signs are assembled into sentences and 
spoken through offline text-to-speech, while the hearing participant replies through offline speech-to-text or the 
keyboard. A context-specific dataset of 60 signs drawn from education, healthcare and workplace scenarios is 
defined, with signer-independent evaluation. The proposed recogniser is expected to reach 94.6% accuracy and 
a 94.3% macro F1-score with 1.18 M parameters and about 26 ms of processing per frame on a standard CPU, 
enabling real-time use without special hardware  

Downloads

Published

2025-06-27

Issue

Section

Articles

How to Cite

Speech-impaired individuals face barriers in communication across education, healthcare & workplaces. (2025). International Journal of Engineering and Science Research, 15(2), 855-865. https://ijesr.org/index.php/ijesr/article/view/1900

Similar Articles

81-90 of 1244

You may also start an advanced similarity search for this article.