An ordinary person can perceive, hear, and react to their environment. However, some individuals lack this critical capability. These individuals, specifically those who are deaf or non-verbal, depend on sign language to communicate. The difficulty arises when they attempt to engage with people who do not understand sign language, complicating interactions. This creates a significant challenge for deaf and mute individuals, particularly in educational, social, and professional settings. The objective of this research is to create a sign language translation system to facilitate communication between people with hearing or speech disabilities and those without, and to assess the system’s effectiveness in interpreting sign language. The initial step involved identifying the most efficient technique for gesture recognition through a review of prior studies. The first stage of testing involved evaluating the tilt sensor’s performance. After assembling all components, the glove’s ability to accurately translate specific alphabets, numbers, and words from Malaysian Sign Language was tested. In the first experiment, it was found that the tilt sensor had to be angled more than 85° to successfully switch the digital state. For the four participants who tested the device, the average accuracy for alphabet translation was 95%, for numbers 93.33%, and for gestures 78.33%. Overall, the glove demonstrated an average accuracy of 89% for translating all gestures. Future improvements to this combination of tilt sensors and accelerometers could include expanding the training and testing datasets, along with utilizing advanced algorithms like the Hidden Markov Model.