Translating SIBI (Sign System for Indonesian Gesture) Gesture-to-Text in Real-Time using a Mobile Device

The SIBI gesture translation framework by Rakun was built using a series of machine learning technologies: MobileNetV2 for feature extraction, Conditional Random Field finding the epenthesis movement frame, and Long Short-Term Memory word classification. This high computational system previously implemented on personal computer system, which lacks portability accessibility. study smartphone an on-device inference method: process is embedded into to provide lower latency zero data usage. then improved parallel multi-inference method, reduced average time 25%. final mobile gesture-to-text achieved accuracy 90.560%, sentence 64%, 20 seconds.