Multiple Acoustic Model-Based Discriminative Likelihood Ratio Weighting for Voice Activity Detection

In this letter, we propose a novel statistical voice activity detection (VAD) technique. The proposed technique employs probabilistically derived multiple acoustic models to effectively optimize the weights on frequency domain likelihood ratios with the discriminative training approach for more accu...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	IEEE signal processing letters 2012-08, Vol.19 (8), p.507-510
Hauptverfasser:	Suh, Youngjoo, Kim, Hoirin
Format:	Artikel
Sprache:	eng
Schlagworte:	Acoustics Discrete Fourier transforms Frequency domain analysis Multiple acoustic models Probabilistic logic Signal to noise ratio Speech statistical voice activity detection Training weighted likelihood ratio
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!