A BAYESIAN APPROACH TO INTER-TASK FUSION FOR SPEAKER RECOGNITION

In i-vector based speaker recognition systems, back-end classifiers are trained to factor out nuisance information and retain only the speaker identity. As a result, variabilities arising due to gender, language and accent ( among many others) are suppressed. Inter-task fusion, in which such metadata information obtained from automatic systems is used, has been shown to improve speaker recognition performance. In this paper, we explore a Bayesian approach towards inter-task fusion. Speaker similarity score for a test recording is obtained by marginalizing the posterior probability of a speaker. Gender and language probabilities for the test audio are combined with speaker posteriors to obtain a final speaker score. The proposed approach is demonstrated for speaker verification and speaker identification tasks on the NIST SRE 2008 dataset. Relative improvements of up to 10% and 8% are obtained when fusing gender and language information, respectively.


Published in:
5786-5790, 1520-6149
Presented at:
In Proceedings of ICASSP 2019, Brighton, ENGLAND
Year:
2019
ISBN:
978-1-4799-8131-1
Keywords:
Laboratories:




 Record created 2019-11-07, last modified 2019-11-07


Rate this document:

Rate this document:
1
2
3
 
(Not yet reviewed)