Název:	Methods of unsupervised adaptation in online speech recognition
Další názvy:	Metody neřízené adaptace online rozpoznávání­ řeči
Autoři:	Machlica, Lukáš Zají­c, Zbyněk Pražák, Aleš
Citace zdrojového dokumentu:	MACHLICA, Lukáš; ZAJÍC, Zbyněk; PRAĹ˝ĂK, Aleš. Methods of unsupervised adaptation in online speech recognition. In: SPECOM 2009 Proceedings. St. Petersburg: Institute for Informatics and Automation of RAS (SPIIRAS), 2009, p. 448-453. ISBN 978-5-8088-0442-5.
Datum vydání:	2009
Nakladatel:	Institute for Informatics and Automation of RAS (SPIIRAS)
Typ dokumentu:	článek article
URI:	http://www.kky.zcu.cz/cs/publications/LukasMachlica_2009_Methodsof http://hdl.handle.net/11025/17044
ISBN:	978-5-8088-0442-5
Klíčová slova:	adaptační­ techniky;pravděpodobnost lineární­ transformace
Klíčová slova v dalším jazyce:	adaptation techniques;likelihood linear transformations
Abstrakt v dalším jazyce:	This paper deals with adaptation techniques based on maximum likelihood linear transformations, which are well suited for the task of on-line recognition. When transcriptions are available before the system starts running, we are speaking about supervised adaptation. In unsupervised adaptation the transcriptions have to be computed in the first pass of the recognition process. This is often the case in on-line recognition, where data are gathered continuously. Because the system does not work perfectly it is suitable to assign a certainty factor (CF) to each of the transcriptions. Only data that transcriptions have high CF are used for the adaptation. In the on-line recognition, the adaptation (in the sense of updating transformation formulas) has to be performed iteratively whenever the amount of recognized data reaches the pre-specified level. When small amount of adaptation data is available, it is suitable to involve regression trees to cluster similar model parameters. It is quite useful to adapt both speech and silence parameters. Because speech and silence have very different characteristics, we have separated them into two different clusters. Presented methods have been tested on short term recordings and results have proved the suitability of the proposed approach.
Práva:	© Lukáš Machlica - Zbyněk Zajíc - Aleš Pražák
Vyskytuje se v kolekcích:	Články / Articles (KKY) Články / Articles (NTIS)

Všechny záznamy v DSpace jsou chráněny autorskými právy, všechna práva vyhrazena.