Skripsi

IMPLEMENTASI MODEL REKOGNISI SUARA MENGGUNAKAN METODE CONVOLUTIONAL RECURRENT NEUTAL NETWORK (CRNN)

---

Penilaian

---

Voice recognition is a process that can determine the identity of the user based on the characteristics of the voice through the words conveyed because every human being has a distinctive sound characteristic. Human voices can be recognized by gender. Recognition of gender or gender based on voice can be detected automatically through the characteristics of the voice. The method used in this research is Convolutional Recurrent Neural Network (CRNN). Based on the tuning process that has been carried out, the parameter values that produce the best output are the distribution of 80% training data and 20% testing data, batch_size of 8, learning-rate of 0.0001 and epoch of 300. The model that has been built produces a Training Accuracy value of 99 ,41% while Testing Accuracy is 99.05%. The average training performance value obtained from each class is 99.3% for the value of Specificity, Precision and Recall, while the FI-Score value is 99.4%. The average Testing performance value obtained from each class is 98.9% for the Specifity, Precision and Recall values, while the FI-Score value is 99%.

---

Availability

Inventory Code	Barcode	Call Number	Location	Status
2107002310	T51006	T510062021	Central Library (Referens)	Available but not for loan - Not for Loan

Detail Information

Series Title

-

Call Number

T510062021

Publisher

Inderalaya : Prodi Sistem Informasi, Fakultas Ilmu Komputer Universitas Sriwijaya., 2021

Collation

xvi, 92 hlm,: ilus.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

005.507

Content Type

-

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Prodi Sistem Informasi
Program Aplikasi dengan Kegunaan Khusus

Specific Detail Info

-

Statement of Responsibility

MURZ

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment