Voice Activity Detection Based on Short-Time Energy and Noise Spectrum Adaptation: Influence Statistics

Expert Impact

Concepts for which they have has direct influence: Voice activity detection , Voice activity , Activity detection , Time energy , Speech signal , Shorttime energy , Short time .

Key People For Voice Activity Detection

Top KOLs in the world
#1
Wonyong Sung
neural networks speech recognition nand flash memory
#2
Jongseo Sohn
voice activity detector soft decision statistical model
#3
Nam Soo Soo Kim
interocular differences retinal illuminance cervical cancers
#4
Jose Carlos Segura
speech recognition histogram equalization evoked potentials
#5
Antonio J Rubio
speech recognition histogram equalization dialogue systems
#6
Javier Ramírez
alzheimer disease early diagnosis intensity normalization

Voice Activity Detection Based on Short-Time Energy and Noise Spectrum Adaptation

Abstract

. On the basis of the short-time energy of speech signal and the efficient method of noise statistics adaptation estimation proposed by Sohn et al., a new high robust voice activity detection (VAD) rule for any kinds of environmental noise is proposed in this paper. The accurate recognition rate of the new method is about five percent higher than that of Sohn's method on average, and also has the same merit of tracking the noise spectrum properly as Sohn's method. A great deal of simulation experiments shows that the new method is an efficient and robust voice activity detector.