ANALISIS SISTEM PENGENAL KATA MENGGUNAKAN SPEKTOGRAM

SYAFRIANTO, SYAFRIANTO (2022) ANALISIS SISTEM PENGENAL KATA MENGGUNAKAN SPEKTOGRAM. Skripsi thesis, Institut Teknologi Dirgantara Adisutjipto.

	Text (ABSTRAK) 18010039_ABSTRAK.pdf Download (12kB)
	Text (BAB I) 18010039_BAB I.pdf Download (92kB)
	Text (DAFTAR PUSTAKA) 18010039_DAFTAR PUSTAKA.pdf Download (9kB)
	Text (FULL SKRIPSI) 18010039.pdf Restricted to Repository staff only Download (1MB)

Abstract

Technological developments are starting rapidly which can help humans in meeting their every need. One of the emerging technologies is voice signal processing. Voice signal processing has been widely applied in various fields such as sound processing, music, biomedical, navigation, telecommunications, and video. In this study, a spectrogram-based word recognition system was built to recognize a word spoken by someone and to obtain system performance then expressed in percentage accuracy. In this study, word recognition analysis will be carried out using the spectrogram method and Euclidean Distance classification, With the aim of recognizing the words spoken by someone with the accuracy of the training data and test data that is proven in the form of the minimum Euclidean Distance results from all experiments. The results of the study indicate that a word recognition system that can identify a word begins with the process of training data and test data. Word recognition made using the spectrogram method and Euclidean Distance classification resulted in an average recognition accuracy of 88.89% which was carried out 9 times testing the test data on 21 training data, with the best recognition accuracy of 100%.

Actions (login required)

View Item

Downloads