PUTRI, AZRA HITA DAHAYU (2026) PIANO CHORD SOUND DETECTION USING A CONVOLUTIONAL NEURAL NETWORK METHOD. Tugas Akhir thesis, University of Technology Yogyakarta.
![[img]](http://eprints.uty.ac.id/style/images/fileicons/text.png) |
Text
5220411226_AZRA HITA DAHAYU PUTRI_ABSTRAK INDONESIA.pdf Download (137kB) |
Abstract
ABSTRACT The piano plays an important role in human life. One essential element of piano playing is the use of chords, which are combinations of notes played simultaneously to produce specific harmonies. Recognizing chord names on the piano is often challenging for both beginners and experienced players, as it requires well-developed auditory skills. Previous research has employed Convolutional Neural Networks (CNNs) to detect major and minor chords with high accuracy; however, real-time detection has not yet been supported. This project proposes using Convolutional Neural Networks (CNNs) and Bidirectional Long Short-Term Memory (BiLSTMs) as a comparison method with MFCC, Delta, and Delta-Delta extraction to detect various types of triad chords, namely major, minor, augmented, and diminished chords. The CNN and BiLSTM models were built using 5280 training data and 960 test data in .wav format from 48 chord categories, with the CNN accuracy of 98.65% and BiLSTM accuracy of 94.90% on the test data. The chord detection system is implemented in Python and designed to predict chords from audio signals recorded with a microphone and uploaded files. Keywords: Chord Detection, Audio Signal Processing, CNN, BiLSTM, MFCC, Piano, Voice Classification.
| Item Type: | Thesis (Skripsi, Tugas Akhir or Kerja Praktek) (Tugas Akhir) |
|---|---|
| Subjects: | T Technology > T Technology (General) > T201 Patents. Trademarks |
| Divisions: | Fakultas Sains Dan Teknologi > S1 Informatika |
| Depositing User: | Kaprodi S1 Informatika UTY |
| Date Deposited: | 07 May 2026 03:01 |
| Last Modified: | 07 May 2026 03:01 |
| URI: | http://eprints.uty.ac.id/id/eprint/19816 |
Actions (login required)
 |
View Item |
