DISTANCE MEASUREMENTS METHOD IN THE DEMITE ENGLISH PRONUNCIATION ASSESSMENT SYSTEM

• Main Author: Muhammad, Alfatih
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/36667
• Institution: Institut Teknologi Bandung

The Demite is an educational game with a horror genre using Augmented Reality and Speech Recognition technology. In this game, players will capture and eliminate ghosts with English pronunciation. Therefore, players will not only capture and eliminate ghosts, but players will also practice their English pronunciation correctly. To find out the exact and correct pronunciation in English, The Demite requires an assessment system to measure the distance in the pronunciation of English words. In this study, we will analyze distance measurement methods Levenshtein Distance, Manhattan Distance and Euclidean Distance, so that we can determine the appropriate method to measure the accuracy of the pronunciation of English words. The parameters used are phonetics, syllables, phonetic length, and word level. The results of the assessment will be accumulated into the value of damage to ghosts. Based on the results of the implementation and testing that has been done, it can be concluded that the use of the Levenshtein Distance distance measurement method allows it to be used because it has high accuracy when tested with the same and similar words, although when tested with different words the average similarity value is 16% not to touch the number 0%. In the process of measuring the distance of the phonetic and syllable parameters, the Manhattan Distance method is more appropriate to use than Euclidean Distance even though at the time of testing the values of these two methods are almost the same, because the demite scoring system requires all the distance parameters while Euclidean Distance takes the shortest distance in the measurement. The Assessment System has been implemented in The Demite game and has been able to calculate the similarity of English words.