Guntur Laksono Putra, . (2024) PENGENALAN EMOSI SUARA MANUSIA MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORK DENGAN KOMBINASI EKSTRAKSI FITUR MFCC DAN GFCC. Skripsi thesis, Universitas Pembangunan Nasional Veteran Jakarta.
![[img]](http://repository.upnvj.ac.id/style/images/fileicons/text.png) |
Text
ABSTRAK.pdf Download (227kB) |
|
Text
AWAL.pdf Download (1MB) |
|
Text
BAB I.pdf Restricted to Repository UPNVJ Only Download (243kB) |
|
Text
BAB II.pdf Restricted to Repository UPNVJ Only Download (713kB) |
|
Text
BAB III.pdf Restricted to Repository UPNVJ Only Download (322kB) |
|
Text
BAB IV.pdf Restricted to Repository UPNVJ Only Download (2MB) |
|
Text
BAB V.pdf Download (229kB) |
|
Text
DAFTAR PUSTAKA.pdf Download (239kB) |
|
Text
RIWAYAT HIDUP.pdf Restricted to Repository UPNVJ Only Download (117kB) |
|
Text
HASIL PLAGIARISME.pdf Restricted to Repository staff only Download (192kB) |
|
Text
ARTIKEL KI.pdf Restricted to Repository staff only Download (998kB) |
Abstract
Emotions are an important factor in interactions between humans, emotions often influence human behavior. Emotions can be recognized through a person's facial expressions, voice characteristics, and body language. Along with the development of technology, humans continue to try to conduct research so that computers can understand human desires and feelings. One technology that is being developed is the ability of computers to recognize emotions through the information contained in the human voice. This research tries to implement the CNN method with combined MFCC and GFCC feature extraction to classify speakers' emotions using the Toronto Emotional Speech Set dataset which consists of 7 emotion classes. The research results show that the use of the MFCC+GFCC feature combination positively influences the performance of the CNN classification model, by achieving an average accuracy value of 98.57%, an average precision of 98.65%, and an average recall of 98.57%. Using a combination of MFCC+GFCC features also increases the average accuracy and recall by 2.62% and increases the average precision by 2.35-2.47% compared to using a single feature. These findings demonstrate the potential of using the combination of MFCC+GFCC features in human voice emotion recognition, with important implications in the development of more sophisticated speech recognition systems responsive to human emotions.
| Item Type: | Thesis (Skripsi) |
|---|---|
| Additional Information: | [No.Panggil: 2010511033] [Pembimbing 1: Jayanta] [Pembimbing 2: Nurul Afifah Arifuddin] [Penguji 1: Ridwan Raafi'udin] [Penguji 2: Iin Ernawati] |
| Uncontrolled Keywords: | Emotion Recognition, Human Voice, MFCC, GFCC, CNN |
| Subjects: | Q Science > Q Science (General) Q Science > QA Mathematics > QA75 Electronic computers. Computer science |
| Divisions: | Fakultas Ilmu Komputer > Program Studi Informatika (S1) |
| Depositing User: | GUNTUR LAKSONO PUTRA |
| Date Deposited: | 05 Sep 2024 01:54 |
| Last Modified: | 05 Sep 2024 01:54 |
| URI: | http://repository.upnvj.ac.id/id/eprint/30257 |
Actions (login required)
 |
View Item |