Detail publikačního výsledku

Non-Negative Factor Analysis of Gaussian Mixture Model Weight Adaptation for Language and Dialect Recognition

BAHARI, M.; DEHAK, N.; VAN HAMME, H.; BURGET, L.; ALI, A.; GLASS, J.

Originální název

Non-Negative Factor Analysis of Gaussian Mixture Model Weight Adaptation for Language and Dialect Recognition

Anglický název

Non-Negative Factor Analysis of Gaussian Mixture Model Weight Adaptation for Language and Dialect Recognition

Druh

Článek WoS

Originální abstrakt

In this research, a non-negative factor analysis approach is developed for GMM weight decomposition and adaptation. This methods allows for a new low-dimensional utterance representation similar to i-vectors.

Anglický abstrakt

In this research, a non-negative factor analysis approach is developed for GMM weight decomposition and adaptation. This methods allows for a new low-dimensional utterance representation similar to i-vectors.

Klíčová slova

Non-negative factor analysis; model adaptation; Gaussian mixture model weight; dialect recognition; language recognition

Klíčová slova v angličtině

Non-negative factor analysis; model adaptation; Gaussian mixture model weight; dialect recognition; language recognition

Autoři

BAHARI, M.; DEHAK, N.; VAN HAMME, H.; BURGET, L.; ALI, A.; GLASS, J.

Rok RIV

2015

Vydáno

01.07.2014

ISSN

2329-9290

Periodikum

IEEE-ACM Transactions on Audio Speech and Language Processing

Svazek

2014

Číslo

7

Stát

Spojené státy americké

Strany od

1117

Strany do

1129

Strany počet

13

URL

BibTex

@article{BUT111657,
  author="Mohamad {Bahari} and Najim {Dehak} and Hugo {Van hamme} and Lukáš {Burget} and Ahmed {Ali} and Jim {Glass}",
  title="Non-Negative Factor Analysis of Gaussian Mixture Model Weight Adaptation for Language and Dialect Recognition",
  journal="IEEE-ACM Transactions on Audio Speech and Language Processing",
  year="2014",
  volume="2014",
  number="7",
  pages="1117--1129",
  doi="10.1109/TASLP.2014.2319159",
  issn="2329-9290",
  url="http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6803908"
}