Farhad Nadi

Sign language is a language created to help the deaf-mute communicate with others. In Indonesia, there are two sign languages: Indonesian sign language system or sistem isyarat bahasa Indonesia (SIBI) and Indonesian sign language or bahasa isyarat Indonesia (BISINDO), but the deaf community prefers to use BISINDO. Deaf-mute people use sign language to communicate, but normal people usually avoid learning it because they feel it's unnecessary for daily communication. This will certainly be a problem for the deaf-mute because people who do not learn sign language will certainly have difficulty understanding what they are trying to convey. Therefore, an application that can translate sign language is needed to help overcome communication problems between the deaf-mute and normal people. This research focuses on BISINDO translation using a temporal convolutional network (TCN) with MediaPipe methods. The amount of data used consists of 2,700 videos representing 27 word-level signs in BISINDO; this number increases to 32,400 after undergoing the data augmentation process. The model built was tested in translating BISINDO at the word level in simulation, and a thorough comparison between TCN and bidirectional long short-term memory (BiLSTM); both of them were running on the MediaPipe platform separately. From the experiment, it is shown that the performance of TCN exceeds BiLSTM with an accuracy rate as high as 88.64%.

Brail is a written mode of communication utilized by individuals with visual impairments to engage in interpersonal exchanges. The braille writing system consists of patterns printed on specialized paper that feature embossed dots. Braille documents enable the visually impaired to acquire knowledge and information exclusively through the application of their sense of contact. Comprehending braille is not a simple undertaking, particularly for the general populace. Because braille is not a required subject in Indonesian education, the majority of the population lacks proficiency in the language. This may therefore result in a minor communication barrier between visually impaired individuals and non-impaired individuals. In order to address this challenge, the present study employs digital image processing via the deep convolutional neural network (DCNN) technique to facilitate comprehension of braille document contents by non-braille speakers. This study employs a deep learning technique that is highly accurate, effective at image processing, and capable of recognizing complex patterns. This study employed the following image processing methods: grayscaling, filtering, contrast enhancement, thresholding, morphological operation, and resizing. Following testing in this investigation, it was determined that the proposed method accurately identifies embossed braille images with a precision of 99.63%.