Corpora: corpora of vowels
Johansson C.L.E. (johanson@let.rug.nl)
Mon, 11 May 1998 10:51:26 +0200 (METDST)
I would like to know how hard it is to get a corpora of classified vowels: i.e.
a corpora where each vowel-representation has a label such as
VowelName_Language_PersonCode where VowelName could be close to IPA
classification and possibly marked for language and the speaker. For the
application/experiment I have in mind it would be necessary to normalise these
vowels into a fixed grid, but it is not a necessary requirement.
If such data is readily available I would be happy to know from where and how
and if applicable the price.
It would also be nice to know how accepted the corpus is and how much other
work have been done on that data (on classification of vowels, language, and
speakers).
Many thanks for any help,
/Christer Johansson (johanson@let.rug.nl)