PENGEMBANGAN APLIKASI PENERJEMAH BAHASA ISYARAT INDONESIA (BISINDO) MENGGUNAKAN METODE LONG-SHORT TERM MEMORY

The Indonesian Sign Language Translator (BISINDO) application is intended for individuals who are deaf and mute assist them in communicating with normal people. Due to the limitations of technology to assist individuals who are deaf and mute, the Indonesian Sign Language Translator application was developed. Additionally, it is hoped that this application will enable individuals who are deaf and mute to learn Indonesian Sign Language in real-time. LSTM generally consists of a cell, input, gate, and forget gate. LSTM is highly suitable for classifying, processing, and making predictions based on time series data. In this research, hand landmark recognition is used as a medium to store finger gestures from user inputs, which will then be saved in the form of coordinate points and stored in a file in csv format. Subsequently, this file will be trained and called back to determine the output.