Sign Language Conversion to Speech with the Application of KNN Algorithm
M. Rajani Shree, N Nadeem Ahmed, Yashvi Panchani, Shreyaa Aravindan, Viraj Jadhav · 2022 Sixth International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC) · 2022
Sign language is mainly used by those who are unable to speak, to communicate. The vast majority of people are not able to understand the Universal Sign Language (unless they have learned it) and due to this lack of knowledge about the language, they have difficulty communicating with mute people. This paper focuses on developing an application that bridges the gap between mute people and the rest of society. It captures gestures made by a person using American sign language (ASL), and converts them into corresponding text and speech in real time. The proposed study will acquire translations of sign language gestures in text, which will be later transformed into audio. In this way, a sign language translator will be developed. Further, Convolutional neural networks (CNNs) are utilized for the detection of different gestures. CNNs are highly effective for solving computer vision problems and can detect the desired features with high degrees of accuracy upon exercising sufficient training. K-NN classification is used to recognize sign language.