VSWarren@aol.com wrote:
> Can anyone please suggest either a program to convert from orthographic to
> phonemic or alternatively a large corpora where phonemic transcriptions are
> given for such a large number of different words.
You can download software that does a pretty good job
of converting text to segmental phonemes from the NIST
website: see http://www.nist.gov/speech/tools/index.htm.
But you should be aware that the output from this is
phonemic underlying forms that often are realized differently
in actual speech; for instance, what usually surfaces as
syllabic consonants are phonemicized as a sequence of
(zero-stressed) schwa plus consonant (as in "button").
A good free lexicon of English is available from CMU;
see http://www.speech.cs.cmu.edu/cgi-bin/pronounce.
In addition, the LDC offers a high-accuracy one, but
it's not free. And Joe Picone & Co. of Mississippi State
are making available lexicons derived from their
re-transcription of the Switchboard corpus along with
phonetic transcriptions from ICSI; see
http://www.isip.msstate.edu/projects/switchboard/index.html.
- Bill F.
This archive was generated by hypermail 2b29 : Tue Nov 14 2000 - 15:51:39 MET