Improved Jacobian Adaptation for Robust Speaker Verification

Jan ANGUITA  Javier HERNANDO  Alberto ABAD  

Publication
IEICE TRANSACTIONS on Information and Systems   Vol.E88-D   No.7   pp.1767-1770
Publication Date: 2005/07/01
Online ISSN: 
DOI: 10.1093/ietisy/e88-d.7.1767
Print ISSN: 0916-8532
Type of Manuscript: LETTER
Category: Speech and Hearing
Keyword: 
Jacobian adaptation,  robust speaker verification,  Frequency Filtering,  

Full Text: PDF(313.3KB)>>
Buy this Article




Summary: 
Jacobian Adaptation (JA) has been successfully used in Automatic Speech Recognition (ASR) systems to adapt the acoustic models from the training to the testing noise conditions. In this work we present an improvement of JA for speaker verification, where a specific training noise reference is estimated for each speaker model. The new proposal, which will be referred to as Model-dependent Noise Reference Jacobian Adaptation (MNRJA), has consistently outperformed JA in our speaker verification experiments.