PENERAPAN EKSTRAKSI FITUR MEL FREQUENCY CEPSTRAL COEFFICIENTS DAN METODE KLASIFIKASI CNN UNTUK IDENTIFIKASI JENIS SUARA MANUSIA BERDASARKAN VOCAL RANGE

• Yazar: Adhiya Delira Yasiin, (Yazar)
• Materyal Türü: Kitap
• Baskı/Yayın Bilgisi: 2023-07-04.
• Konular: Thesis; NonPeerReviewed
• Online Erişim: Link Metadata
• Notes: http://repository.upnvj.ac.id/25240/13/ABSTRAK.pdf; http://repository.upnvj.ac.id/25240/14/AWAL.pdf; http://repository.upnvj.ac.id/25240/3/BAB%20I.pdf; http://repository.upnvj.ac.id/25240/4/BAB%20II.pdf; http://repository.upnvj.ac.id/25240/5/BAB%20III.pdf; http://repository.upnvj.ac.id/25240/6/BAB%20IV.pdf; http://repository.upnvj.ac.id/25240/7/BAB%20V.pdf; http://repository.upnvj.ac.id/25240/8/DAFTAR%20PUSTAKA.pdf; http://repository.upnvj.ac.id/25240/9/DAFTAR%20RIWAYAT%20HIDUP.pdf; http://repository.upnvj.ac.id/25240/10/LAMPIRAN.pdf; http://repository.upnvj.ac.id/25240/11/HASIL%20PLAGIARISME.pdf; http://repository.upnvj.ac.id/25240/12/ARTIKEL%20KI.pdf
• Keywords: M Music; QA75 Electronic computers. Computer science; QA76 Computer software

The choir is a collection of several people and combines various types of sound or timbre into a unified whole. From this understanding, it can be seen that determining the types of voices in a choir is very important for the composition of the choir itself. However, the determination of the type of voice is usually still done manually, namely with the help of experts in the field of music or vocal trainers. Prospective choir members who have never entered the world of choirs will usually be tested for vocals and the type of voice will be determined by looking at the vocal range of the prospective new member himself. From this habit, the determination of the type of voice becomes very dependent on the vocal trainer and the musical instrument used. Therefore, this research was made in the form of a machine learning model that aims to see the accuracy or performance of feature extraction and classification methods in recognizing voice types based on vocal range. This study uses the Mel Frequency Cepstral Coefficients (MFCC) feature as feature extraction and the Convolutional Neural Network (CNN) method as a sound type classification method. This research was able to produce a system model that has good performance in distinguishing soprano, alto, tenor and bass classes with validation accuracy reaching 95%, and testing accuracy reaching 95% for the Voice Classification model.h5 with high precision, recall, and F1-score for each class.