SpecAugment Impact on Automatic Speaker Verification System

MUHAMMAD YUSUF FAISAL

Informasi Dasar

54 kali
20.04.1036
006.3
Karya Ilmiah - Skripsi (S1) - Reference

An automatic speaker verification (ASV) is one of the challenging problem in speech processing since there are so many models of machine learnings those capable of synthesizing a fake speech from a given text. This paper discusses the impact of SpecAugment to state of the art methods such as Gaussian Mixture Models (GMM) and Deep Neural Networks (DNNs). Some experiments on a speech dataset sampled from the ASVSpoof2019, which is specially made to tackle the threat of spoofing, show that GMM produces an Equal Error Rate (EER) of 19.0% that is better than the DNNs system with EER of 24.0%. However, after combining with a traditional augmentation technique, the DNN gives a better EER of 15.3% than GMM with EER of 15.7%.

Subjek

ARTIFICIAL INTELLIGENCE
 

Katalog

SpecAugment Impact on Automatic Speaker Verification System
 
 
Indonesia

Sirkulasi

Rp. 0
Rp. 0
Tidak

Pengarang

MUHAMMAD YUSUF FAISAL
Perorangan
SUYANTO, NIKEN DWI WAHYU CAHYANI
 

Penerbit

Universitas Telkom, S1 Informatika
Bandung
2020

Koleksi

Kompetensi

  • MUG1E3 - ALJABAR LINEAR
  • MUG2D3 - PROBABILITAS DAN STATISTIKA

Download / Flippingbook

 

Ulasan

Belum ada ulasan yang diberikan
anda harus sign-in untuk memberikan ulasan ke katalog ini