EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Analisis Pengaruh Segmentasi Temporal Sinyal Audio terhadap Akurasi Klasifikasi Nada Rindik Menggunakan Algoritma SVM dan CNN

I Gede Bintang Arya Budaya, I Gede Putra Mas Yusadara, Putu Setyarini, Komang Gita Meiliana

Jurnal Informatika: Jurnal Pengembangan IT · 16.10.2026 · 11 (3) · 849-857

DOI: 10.30591/jpit.v11i3.48 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Jurnal Informatika: Jurnal Pengembangan IT, online erschienen am 16.10.2026. Verfasst von I Gede Bintang Arya Budaya und 3 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The traditional Balinese Rindik musical instrument has unique acoustic characteristics in the form of complex bamboo resonance, posing challenges in automatic pitch recognition due to timbre similarities. This study aims to classify 11 slendro notes in Rindik and analyze the significance of signal temporal segmentation's influence on classification accuracy. The proposed method applies audio signal cutting into three acoustic phases: Attack, Sustain, and Release. Feature extraction is performed using 40 Mel-Frequency Cepstral Coefficients (MFCC). Classification is executed using two comparison algorithms, namely Support Vector Machine (SVM) and 1DConvolutional Neural Network (1D-CNN), on an augmented dataset with an 80:20 data ratio. Experimental results show that SVM consistently outperforms CNN across all testing scenarios. A specific finding in this study is that the use of the Sustain phase (middle) produces the highest accuracy of 100%, which is proven to be more effective than both the Attack and release phase (97.27%) or the use of the intact signal (Full Envelope) with an accuracy of 99.09%. This indicates that the most stable spectral information and primary pitch discriminators in the Rindik instrument lie in the bamboo resonance phase following the initial transient

Zitieren

APA

Budaya, I. G. B. A., Yusadara, I. G. P. M., Setyarini, P., & Meiliana, K. G. (2026). Analisis Pengaruh Segmentasi Temporal Sinyal Audio terhadap Akurasi Klasifikasi Nada Rindik Menggunakan Algoritma SVM dan CNN. Jurnal Informatika: Jurnal Pengembangan IT, 11(3), 849-857. https://doi.org/10.30591/jpit.v11i3.48

Vancouver

Budaya IGBA, Yusadara IGPM, Setyarini P, Meiliana KG. Analisis Pengaruh Segmentasi Temporal Sinyal Audio terhadap Akurasi Klasifikasi Nada Rindik Menggunakan Algoritma SVM dan CNN. Jurnal Informatika: Jurnal Pengembangan IT. 2026;11(3):849-857. doi:10.30591/jpit.v11i3.48

BibTeX
@article{Budaya2026,
  author = {Budaya, I Gede Bintang Arya and Yusadara, I Gede Putra Mas and Setyarini, Putu and Meiliana, Komang Gita},
  title = {Analisis Pengaruh Segmentasi Temporal Sinyal Audio terhadap Akurasi Klasifikasi Nada Rindik Menggunakan Algoritma SVM dan CNN},
  journal = {Jurnal Informatika: Jurnal Pengembangan IT},
  year = {2026},
  volume = {11},
  number = {3},
  pages = {849--857},
  doi = {10.30591/jpit.v11i3.48}
}
Kostenlos starten →