Loading...

Speaker models reduction for optimized telephony text-prompted speaker verification

Kalantari, E ; Sharif University of Technology | 2015

584 Viewed
  1. Type of Document: Article
  2. DOI: 10.1109/CCECE.2015.7129497
  3. Publisher: Institute of Electrical and Electronics Engineers Inc , 2015
  4. Abstract:
  5. In this article a new scheme is proposed to use mean supervector in text-prompted speaker verification system. In this scheme, for each month name a subsystem is constructed and a final score based on passphrase is computed by the combination of the scores of these subsystems. Results from the telephony dataset of Persian month names show that the proposed method significantly reduces EER in comparison with the-State-of-the-art State-GMM-MAP method. Furthermore, it is shown that based on training set and testing set we can use 12 models per speaker instead of 220. Therefore, this scheme reduces EER and computational burden. In addition, the use of HMM instead of GMM as words' model improves the performance of the system. In the best case, EER is reduced by 32.3%
  6. Keywords:
  7. Computational burden ; Map methods ; Speaker model ; Speaker verification ; Speaker verification system ; State of the art ; Supervector ; Training sets ; Speech recognition
  8. Source: Canadian Conference on Electrical and Computer Engineering, 3 May 2015 through 6 May 2015 ; Volume 2015-June, Issue June , May , 2015 , Pages 1470-1474 ; 08407789 (ISSN)
  9. URL: http://ieeexplore.ieee.org/document/7129497