DocumentCode
2150838
Title
Speaker Identification Using GMM with Embedded AANN
Author
Cunbao Chen ; Li Zhao ; Yan Zhao
Author_Institution
Sch. of Inf. Sci. & Eng., Southeast Univ., Nanjing, China
fYear
2009
fDate
20-22 Sept. 2009
Firstpage
1
Lastpage
4
Abstract
This paper proposes a modified gaussian mixed model (GMM) with an embedded auto-associate neural network (AANN). It integrates the merits of GMM and AANN. GMM and AANN are trained as a whole by means of maximum likelihood. In the process of training, the parameter of GMM and AANN are updated alternately. AANN reshapes the distribution of the data and improves the similarity of the data in one class. Experiments show that the proposed system improves accuracy rate against baseline GMM at all SNR, maximum to 19%.
Keywords
Gaussian processes; maximum likelihood estimation; neural nets; speaker recognition; Gaussian mixed model; auto-associate neural network; automatic speaker recognition; maximum likelihood; speaker identification; Covariance matrix; Hidden Markov models; Information science; Intelligent networks; Neural networks; Probability density function; Speaker recognition; Support vector machines; Testing; Training data;
fLanguage
English
Publisher
ieee
Conference_Titel
Management and Service Science, 2009. MASS '09. International Conference on
Conference_Location
Wuhan
Print_ISBN
978-1-4244-4638-4
Electronic_ISBN
978-1-4244-4639-1
Type
conf
DOI
10.1109/ICMSS.2009.5303933
Filename
5303933
Link To Document