Paper Info
Title
Extracting speech features from human speech like noise
Abstract
Human speech-like noise (HSLN) is a kind of bubble noise generated by superimposing independent speech signals typically more than one thousand times. Since the basic feature of HSLN varies from that of overlapped speech to stationary noise, keeping long time spectra in the same shape, we investigate perceptual discrimination of speech from stationary noise and its acoustic correlates using HSLN of various numbers of superposition. First we confirm the perceptual score, i.e. how much the HSLN sounds like stationary noise, and that the number of superpositions of HSLN is proportional by subjective tests. Then, we show that the amplitude distribution of the difference signal of HSLN approaches the Gaussian distribution from the Gamma distribution as the number of superpositions increase. The other subjective test to perceive three HSLN of different dynamic characteristics clarifies that the temporal change of spectral envelope plays an important roll in discriminating speech from noise
Year
DOI
Venue
1996
10.1109/ICSLP.1996.607143
Spoken Language, 1996. ICSLP 96. Proceedings., Fourth International Conference
Keywords
Field
DocType
Gaussian distribution,feature extraction,gamma distribution,noise,spectral analysis,speech recognition,Gamma distribution,Gaussian distribution,HSLN,acoustic correlates,amplitude distribution,bubble noise,difference signal,human speech like noise,independent speech signals,overlapped speech,perceptual discrimination,perceptual score,spectral envelope,speech feature extraction,stationary noise
Value noise,Colors of noise,Noise floor,Noise measurement,Pattern recognition,White noise,Speech recognition,Artificial intelligence,Gaussian noise,Noise,Mathematics,Gradient noise
Conference
Volume
ISBN
Citations 
1
0-7803-3555-4
8
PageRank 
References 
Authors
1.28
1
4
Name
Order
Citations
PageRank
Daisuke Kobayashi181.28
Shoji Kajita214721.92
Kazuya Takeda31301195.60
Fumitada Itakura443167.73