Re: Corpora: TIMIT syllabification

From: wmfisher@attglobal.net
Date: Tue Nov 27 2001 - 16:17:13 MET

  • Next message: Patrick Ruch: "Corpora: First CFP for NLPBA'02"

    Fabio -

      You can download vanilla C software for syllabification from:

    http://www.nist.gov/speech/tools/index.htm

      look for "syllabification software" under "Language Technology Tools".
     You should be aware that the syllabifications produced are a function of
    the phonemic transcription and a level of formality of the speech, and so
    you'll get several alternate outputs in some cases if you run it on only
    the transcription. They're also based on Dan Kahn's theory, and so
    might be proven wrong by acoustic evidence. (It would be a real contribution
    if someone would make syllabification judgements on this corpus based
    on such real evidence.)

    - Bill F.

    ----- Original Message -----
    From: "Fabio Tamburini" <tamburin@cilta.unibo.it>
    To: <corpora@hd.uib.no>
    Sent: Thursday, November 22, 2001 12:45 PM
    Subject: Corpora: TIMIT syllabification

    > Hi
    > I need to identify the syllable in the phone transcription of the
    > sentences in the TIMIT database.
    >
    > Is there some program for doing that?
    > Someone has already done such kind of work on the TIMIT database?
    >
    > Thanks for any help
    >
    > Fabio Tamburini
    > CILTA - University of Bologna
    >
    >



    This archive was generated by hypermail 2b29 : Tue Nov 27 2001 - 16:38:44 MET