Lexicon optimization for Dutch speech recognition in spoken document retrieval

    Research output: Chapter in Book/Report/Conference proceedingConference contributionProfessional

    11 Downloads (Pure)

    Abstract

    In this paper, ongoing work concerning the language modelling and lexicon optimization of a Dutch speech recognition system for Spoken Document Retrieval is described: the collection and normalization of a training data set and the optimization of our recognition lexicon. Effects on lexical coverage of the amount of training data, of decompounding compound words and of different selection methods for proper names and acronyms are discussed.
    Original languageUndefined
    Title of host publicationProceedings of Eurospeech 2001 - Scandinavia
    EditorsP. Dalsgaard, B. Lindberg, H. Benner
    Pages1085-1088
    Publication statusPublished - 2001

    Keywords

    • METIS-203287
    • IR-36866

    Cite this

    Ordelman, R. J. F., van Hessen, A. J., & de Jong, F. M. G. (2001). Lexicon optimization for Dutch speech recognition in spoken document retrieval. In P. Dalsgaard, B. Lindberg, & H. Benner (Eds.), Proceedings of Eurospeech 2001 - Scandinavia (pp. 1085-1088)