DocumentCode :
573238
Title :
A fast two-level Speaker Identification method employing sparse representation and GMM-based methods
Author :
Zeinali, Hossein ; Sameti, Hossein ; Khaki, Hossein ; BabaAli, Bagher
Author_Institution :
Dept. of Comput. Eng., Sharif Univ. of Technol., Tehran, Iran
fYear :
2012
fDate :
2-5 July 2012
Firstpage :
45
Lastpage :
48
Abstract :
In large population Speaker Identification (SI), computation time has become one of the most important issues in recent real time systems. Test computation time depends on the cost of likelihood computation between test features and registered speaker models. For real time application of speaker identification, system must identify an unknown speaker quickly. Hence the conventional SI methods cannot be used. In this paper, we propose a two-step method that utilizes two different identification methods. In the first step we use Nearest Neighbor method to decrease the search space. In the second step we use GMM-based SI methods to specify the target speaker. We achieved 3.5× speed-ups without any loss of accuracy using the proposed method. If the number of best speaker is reduced, the Identification accuracy decreases. So, there is a trade-off between accuracy and speed-up.
Keywords :
Gaussian processes; signal representation; speaker recognition; GMM-based SI method; likelihood computation; nearest neighbor method; real time application; search space; sparse representation; speaker model; target speaker; test computation time; two-level speaker identification method; Accuracy; Computational modeling; Silicon; Sociology; Statistics; Training; Vectors;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Information Science, Signal Processing and their Applications (ISSPA), 2012 11th International Conference on
Conference_Location :
Montreal, QC
Print_ISBN :
978-1-4673-0381-1
Electronic_ISBN :
978-1-4673-0380-4
Type :
conf
DOI :
10.1109/ISSPA.2012.6310594
Filename :
6310594
Link To Document :
بازگشت