Wearable augmentative and alternative communication device for paralysis victims using Brute Force Algorithm for pattern recognition

In this study, the aim is to design and build a wearable Augmentative and Alternative Communication (AAC) device that converts discrete breathing patterns into pre-defined words to help paralysis victims communicate their needs. In line with this, classification of breath signals into heavy and soft blow to create discrete breathing patterns were investigated. Paralysis victims in a certain age group with no loss of cognitive function shall be the concern of study. In the design that there were no possible threshold values that can be set as constants to classify breath signals into soft and heavy blow. The amplitude of the heavy and soft blow differed in every person based from the conducted statistical test producing a wide deviation, possible overlapping and making no possible threshold value. With this in mind, the researchers developed a program using Brute Force Algorithm and asks the user to register his/her soft and heavy blow during initialization and use it as the threshold upon usage. The accuracy of the device was tested using Standard Error of the Mean and scored 15.45 ± 0.1141 score out of 16.