Sign Language Recognition and Translation Systems: A Comprehensive Review with Special Focus on Malayalam Sign Language (MSL)

Abstract: The main form of communication for those with speech and hearing problems is sign language. Recent developments in artificial intelligence, computer vision, and machine learning have greatly expedited the development of automated sign language detection and translation systems. These technologies use sign-to-text, sign-to-speech, speech-to-sign, and text-to-sign translation methods to close the communication gap between hearing-impaired people and the hearing community. Malayalam Sign Language (MSL), which was legally adopted in Kerala in 2021, has received comparatively less attention than American Sign Language (ASL), Arabic Sign Language, and other extensively used sign languages. With an emphasis on machine learning, deep learning, transfer learning, graph convolutional networks, and multimodal communication frameworks, this study offers a thorough analysis of current advancements in sign language identification and translation systems. The review highlights current issues and potential future research areas while analyzing present approaches, datasets, methodologies, and performance measures. This study gives special attention to Malayalam Sign Language detection and the necessity for effective, scalable, and real-time assistive communication technology.

Keywords: Sign Language Recognition, Malayalam Sign Language, American Sign Language, Deep Learning, Machine Learning, Assistive Technology