Publications (5)0 Total impact
-
[show abstract]
[hide abstract]
ABSTRACT: Whereas nowadays within-word co-articulation effects are usually sufficiently dealt with in automatic speech recognition, this is not always the case with phrase level co-articulation effects (PLC). This paper describes a first approach in dealing with phrase level co-articulation by applying these rules on the reference transcripts used for training our recogniser and by adding a set of temporary PLC phones that later on will be mapped on the original phones. In fact we temporarily break down acoustic context into a general and a PLC context. With this method, more robust models could be trained because phones that are confused due to PLC effects like for example /v/-/f/ and /z/-/s/, receive their own models. A first attempt to apply this method is described.
10/2003;
-
[show abstract]
[hide abstract]
ABSTRACT: This paper describes a first approach to improve recognition performance of our hybrid large vocabulary continuous speech recogniser for Dutch by using co-articulation rules on the phrase level. By applying these rules on the reference transcripts used for training the recogniser and by adding a set of special temporary phones that later on will be mapped on the original phones, more robust models of phones that are typically confused a lot in speech recognition like /v/-/f/ and /s/-/z/, could be trained.
10/2003;
-
[show abstract]
[hide abstract]
ABSTRACT: This paper describes the first large-scale evaluation of information retrieval systems using Dutch documents and queries. We describe in detail the characteristics of the Dutch test data, which is part of the official CLEF multilingual texttual database, and give an overview of the experimental results of companies and research institutions that participated in the first official Dutch CLEF experiments. Judging from these experiments, the handling of language-specific issues of Dutch, like for instance simple morphology and compound nouns, significantly improves the performance of information retrieval systems in many cases. Careful examination of the test collection shows that it serves as a reliable tool for the evaluation of information retrieval systems in the future.
03/2002;
-
[show abstract]
[hide abstract]
ABSTRACT: Whereas nowadays within-word co-articulation effects are usually sufficiently dealt with in automatic speech recognition, this is not always the case with phrase level co-articulation effects (PLC). This paper describes a first approach in dealing with phrase level co-articulation by applying these rules on the reference transcripts used for training our recogniser and by adding a set of temporary PLC phones that later on will be mapped on the original phones. In fact we temporarily break down acoustic context into a general and a PLC context. With this method, more robust models could be trained because phones that are confused due to PLC effects like for example /v/-/f/ and /z/-/s/, receive their own models. A first attempt to apply this method is described. 1. INTRODUCTION The DRUID 1 project (Document Retrieval Using Intelligent Disclosure), a collaboration of CTIT 2 /University of Twente, TNO 3 and CWI 4 , aims at the development of tools for the indexing of multimedia c...
04/1999;
-
[show abstract]
[hide abstract]
ABSTRACT: This paper describes the first large-scale evaluation of information retrieval systems using Dutch documents and queries. We describe in detail the characteristics of the Dutch test data, which is part of the official CLEF multilingual texttual database, and give an overview of the experimental results of companies and research institutions that participated in the first official Dutch CLEF experiments. Judging from these experiments, the handling of language-specific issues of Dutch, like for instance simple morphology and compound nouns, significantly improves the performance of information retrieval systems in many cases. Careful examination of the test collection shows that it serves as a reliable tool for the evaluation of information retrieval systems in the future.
