Pre-processing input text: improving pronunciation for the fluent Dutch text-to-speech systems

Rik Jansen, Adrianus J. van Hessen, Louis C.W. Pols

    Research output: Contribution to conferencePaperpeer-review

    43 Downloads (Pure)

    Abstract

    To improve pronunciation of the Fluent Dutch Text-To-Speech Synthesiser, two pre-processors were built that try to detect problematic cases in input texts and solve these automatically if possible. One pre-processor examines the pronounceability of surnames and company names by checking whether their initial and final two-letter combinations can be handled by the grapheme-to-phoneme rules of the Fluency TTS system, and correcting those automatically when and if possible. Also, common disambiguous abbreviations are properly expanded. The second pre-processor tries to realise pronounceable forms for numbers that do not have a straightforward pronunciation. Structural and contextual information is used in an attempt to determine to what category a number belongs, and each number is expanded according to the pronunciation conventions of its category. It can be said that these pre-processors are a useful aid in offline pronounceability examination (for names) and improvement of performance at run-time (for numbers), although ambiguity and redundancy in the input text illustrate the need for semantic and syntactic parsing to approach human text interpretation skills.
    Original languageEnglish
    Number of pages8
    Publication statusPublished - 2 Feb 1999
    Event14th International Congress of Phonetic Sciences, ICPhS 1999 - San Francisco, United States
    Duration: 1 Aug 19997 Aug 1999
    Conference number: 14
    https://www.internationalphoneticassociation.org/icphs/icphs1999

    Conference

    Conference14th International Congress of Phonetic Sciences, ICPhS 1999
    Abbreviated titleICPhS
    CountryUnited States
    CitySan Francisco
    Period1/08/997/08/99
    Internet address

    Keywords

    • METIS-119612
    • IR-89649

    Fingerprint

    Dive into the research topics of 'Pre-processing input text: improving pronunciation for the fluent Dutch text-to-speech systems'. Together they form a unique fingerprint.

    Cite this