Sixth European Conference on Speech Communication and Technology
(EUROSPEECH'99)

Budapest, Hungary
September 5-9, 1999

Maximum Likelihood Eigenspace and MLLR for Speech Recognition in Noisy Environments

Patrick Nguyen (1,2), Christian Wellekens (1), Jean-Claude Junqua (2)

(1) Institut Eurécom, Sophia-Antipolis, France
(2) Speech Technology Laboratory, Santa Barbara, CA, USA

A technique for rapid speaker adaptation, called eigenvoices, was introduced recently. The key idea is to confine models in a very low-dimensional lin-ear vector space. This space summarizes a prioriknowl-edge that we have about speaker models. In many practical systems, however, there is a mismatch between the conditions in which the training data were collected and test conditions: prior knowledge becomes improper. Furthermore, prior statistics or models of this mismatch may not be available. We expose two key results: first, we use a maximum-likelihood estimator of prior infor-mation in matched conditions, called MLES, leading to an improvement of adaptation by a relative 14%, and second, we show how we can apply a blind scheme for learning noise, MLLR, achieving an additional 7:7% relative improvement in noisy conditions.


Full Paper (PDF)   Gnu-Zipped Postscript

Bibliographic reference.  Nguyen, Patrick / Wellekens, Christian / Junqua, Jean-Claude (1999): "Maximum likelihood eigenspace and MLLR for speech recognition in noisy environments", In EUROSPEECH'99, 2519-2522.