Skip to content

Publications

Detail of publication

Citation

Trmal Jan and Zelinka Jan and Luděk Müller : On Speaker Adaptive Training of Artificial Neural Networks . Proceedings of Int. Conf. Interspeech 2010, 2010.

Download PDF

PDF

Abstract

In the paper we present two techniques improving the recognition accuracy of multilayer perceptron neural networks (MLP ANN) by means of adopting Speaker Adaptive Training. The use of the MLP ANN, usually in combination with the TRAPS parametrization, includes applications in speech recognition tasks, discriminative features production for GMM-HMM and other. In the first SAT experiments, we used the VTLN as a speaker normalization technique. Moreover, we developed a novel speaker normalization technique called Minimum Error Linear Transform (MELT) that resembles the cMLLR/fMLLR method \cite{gales96variance} with respect to the possible application either on the model or features. We tested these two methods extensively on telephone speech corpus SpeechDat-East. The results obtained in these experiments suggest that incorporation of SAT into MLP ANN training process is beneficial and depending on the setup leads to significant decrease of phoneme error rate (3% -- 8% absolute, 12% -- 25% relative).

Detail of publication

Title: On Speaker Adaptive Training of Artificial Neural Networks
Author: Trmal Jan ; Zelinka Jan ; Luděk Müller
Language: English
Date of publication: 1 Sep 2010
Year: 2010
Type of publication: Papers in proceedings of reviewed conferences
Title of journal or book: Proceedings of Int. Conf. Interspeech 2010
Date: 27 Sep 2010 - 30 Sep 2010
/ 2010-10-19 13:39:14 /

BibTeX

@INPROCEEDINGS{TrmalJan_2010_OnSpeakerAdaptive,
 author = {Trmal Jan and Zelinka Jan and Lud\v{e}k M\"{u}ller},
 title = {On Speaker Adaptive Training of Artificial Neural Networks},
 year = {2010},
 journal = {Proceedings of Int. Conf. Interspeech 2010},
 url = {http://www.kky.zcu.cz/en/publications/TrmalJan_2010_OnSpeakerAdaptive},
}