CEPSTRUM PARAMETERS FOR HUMAN VOICE RECOGNITION

Widyatama Repository

CEPSTRUM PARAMETERS FOR HUMAN VOICE RECOGNITION

Show full item record

Title: CEPSTRUM PARAMETERS FOR HUMAN VOICE RECOGNITION
Author: Bhaskoro, Susetyo Bagas
Abstract: Several information which often obtained by the human usually through a communication interview. The Human behavior in communicating will produce some ability basic as follow: (1) face to face process, able to analyze something rapidly in remembering type of sound and identity from the speaker; and (2) the process without face to face, able to hear, remember as well as good making decision from identity of the speaker. In this research, make recognition process and identification that previously to be done by the human through quick training process, precise and easy with communicated system. Today the process can be made through computer, started with the process how get characteristic from the sound of speaker and furthermore conclude the result that agree with its training data. The characteristic that found by each the speaker is cepstrum, value of cepstrum that will be stored and made the reference for each speaker on training and examination phases. The examination is made to 12 speakers with the different genders, with each speaker give sound sample as much as 20 times. The examination result of sound recognition process shows the success as big as 70% for Offline examination and 74.167% for real-time examination with total data examination as much as 10 times to each speaker.
URI: http://repository.widyatama.ac.id/xmlui/handle/123456789/3581
Date: 2013-11-21


Files in this item

Files Size Format View
26Susetyo bagas.pdf 364.6Kb PDF View/Open

This item appears in the following Collection(s)

Show full item record