Rafliansyah, Royan Hisyam (2024) Klasifikasi Suara Instrumen Musik Tiup Menggunakan Metode Convolutional Neural Network. Undergraduate thesis, UPN Veteran Jawa Timur.
![[img]](https://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (Cover)
20081010130-cover.pdf Download (1MB) |
|
Text (Bab 1)
20081010130-bab1.pdf Download (608kB) |
|
Text (Bab 2)
20081010130.-bab2.pdf Restricted to Repository staff only until 4 June 2026. Download (1MB) |
|
Text (Bab 3)
20081010130.-bab3.pdf Restricted to Repository staff only until 4 June 2026. Download (918kB) |
|
Text (Bab 4)
20081010130.-bab4.pdf Restricted to Repository staff only until 4 June 2026. Download (2MB) |
|
Text (Bab 5)
20081010130.-bab5.pdf Download (588kB) |
|
Text (Daftar Pustaka)
20081010130.-daftarpustaka.pdf Download (480kB) |
|
Text (Lampiran)
20081010130.-lampiran.pdf Restricted to Repository staff only until 4 June 2026. Download (447kB) |
Abstract
Music is an art that has been inseparable from human life for centuries. In the art of music itself, there are many types of musical instruments ranging from string instruments, wind instruments, percussion instruments, string instruments, and many other types of instruments. One of the musical instruments that has its own characteristics is wind instruments. There are many types of wind instruments such as saxophone, clarinet, trumpet, and many others. This research uses the Convolutional Neural Network (CNN) algorithm to classify the sound of wind instruments. In the feature extraction process, various techniques are used such as Spectral Contrast, Spectral Bandwidth, Spectral Rolloff, Zero-Crossing Rate, Mel-Frequency Cepstral Coefficients (MFCC), Mel-Scaled Spectrogram, Harmony, and Chroma. The results of this study show that in the first test, the prediction accuracy reached 84%. In the second test, the accuracy decreased to 80%. The third test resulted in an accuracy of 79%, while the fourth test achieved an accuracy of 77%. In the fifth test, the prediction accuracy was recorded at 77%, and in the sixth test also obtained a prediction accuracy of 78%. Each test was conducted using different feature extraction datasets and Convolutional Neural Network models to determine the best results.
| Item Type: | Thesis (Undergraduate) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||||||
| Subjects: | T Technology > T Technology (General) | ||||||||||||
| Divisions: | Faculty of Computer Science > Departemen of Informatics | ||||||||||||
| Depositing User: | Royan Hisyam Rafliansyah | ||||||||||||
| Date Deposited: | 04 Jun 2024 06:45 | ||||||||||||
| Last Modified: | 04 Jun 2024 06:45 | ||||||||||||
| URI: | https://repository.upnjatim.ac.id/id/eprint/24050 |
Actions (login required)
 |
View Item |