Scholarlink

Bengali Sign Language (BdSL) is a vital communication tool for the Deaf community in Bangladesh, relying on hand, facial, and body movements. This research leverages advanced machine learning and deep learning to develop a real-time BdSL recognition system using a Kaggle-sourced dataset of 1200 RGB images resized to 128 × 128 pixels. Six deep learning models were tested, with MobileNetV2 achieving the highest accuracy of 99.89% on training data and 95.85% on testing data. The system provides a practical solution for enhancing accessibility and communication for the deaf community in Bangladesh. This work is substantially different than the existing work where each sample contains complete sentences, and the research recognizes a specific target at the character level. It includes a bilingual text-to-speech conversion facility. The system converts recognized signs to English and Bengali and widens the accessibility and overcomes language barriers.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Real-Time Bengali Sign Language Recognition and Bilingual Text-to-Speech Conversion Using Machine Learning Techniques

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Yeastic Hassan Zihan,
  • Iffaat Ara Mehnaz,
  • Anisha Shawana Sharif,
  • Nusrat Jahan Tisha,
  • Rashedur M. Rahman

摘要

Bengali Sign Language (BdSL) is a vital communication tool for the Deaf community in Bangladesh, relying on hand, facial, and body movements. This research leverages advanced machine learning and deep learning to develop a real-time BdSL recognition system using a Kaggle-sourced dataset of 1200 RGB images resized to 128 × 128 pixels. Six deep learning models were tested, with MobileNetV2 achieving the highest accuracy of 99.89% on training data and 95.85% on testing data. The system provides a practical solution for enhancing accessibility and communication for the deaf community in Bangladesh. This work is substantially different than the existing work where each sample contains complete sentences, and the research recognizes a specific target at the character level. It includes a bilingual text-to-speech conversion facility. The system converts recognized signs to English and Bengali and widens the accessibility and overcomes language barriers.