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.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Jan ANGUITA, Javier HERNANDO, Alberto ABAD, "Improved Jacobian Adaptation for Robust Speaker Verification" in IEICE TRANSACTIONS on Information,
vol. E88-D, no. 7, pp. 1767-1770, July 2005, doi: 10.1093/ietisy/e88-d.7.1767.
Abstract: 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.
URL: https://globals.ieice.org/en_transactions/information/10.1093/ietisy/e88-d.7.1767/_p
Copy
@ARTICLE{e88-d_7_1767,
author={Jan ANGUITA, Javier HERNANDO, Alberto ABAD, },
journal={IEICE TRANSACTIONS on Information},
title={Improved Jacobian Adaptation for Robust Speaker Verification},
year={2005},
volume={E88-D},
number={7},
pages={1767-1770},
abstract={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.},
keywords={},
doi={10.1093/ietisy/e88-d.7.1767},
ISSN={},
month={July},}
Copy
TY - JOUR
TI - Improved Jacobian Adaptation for Robust Speaker Verification
T2 - IEICE TRANSACTIONS on Information
SP - 1767
EP - 1770
AU - Jan ANGUITA
AU - Javier HERNANDO
AU - Alberto ABAD
PY - 2005
DO - 10.1093/ietisy/e88-d.7.1767
JO - IEICE TRANSACTIONS on Information
SN -
VL - E88-D
IS - 7
JA - IEICE TRANSACTIONS on Information
Y1 - July 2005
AB - 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.
ER -