[comp.archives] [ai] Rhosettastone: A speech recognition and synthesis benchmark

eugene@eos.UUCP (Eugene Miya) (04/06/90)

Archive-name: miya-human-speech/05-Apr-90
Original-posting-by: eugene@eos.UUCP (Eugene Miya)
Original-subject: Rhosettastone: A speech recognition and synthesis benchmark
Archive-site: aurora.arc.nasa.gov [128.102.21.1]
Archive-directory: pub
Reposted-by: emv@math.lsa.umich.edu (Edward Vielmetti)

This is my yearly call for interesting examples of human speech.
The examples should be of interest to those doing speech
synthesis or recognition.  I collect the examples so that a naive
public isn't so easily fooled by tricksters.  For instance the following
sound very similar:
	How to recognize speech.
	How to wreck a nice beach.
I would especially like examples of speech in foreign languages.
Please send me mail.  Do not post.  The contents are stored for
anonymous FTP on aurora in pub.  I check for duplicates.   Please
feel free to forward this note to the foreign language group of your
choice.

--e.n. miya, NASA Ames Research Center, eugene@eos.arc.nasa.gov
  {uunet,mailrus,most gateways}!ames!eugene