The realm of speech synthesis/recognition uses rules created by the US Navy to convert phonemes to text. Surely this means the systems will output odd text when they hear some proper English? (Although thinking about it, not many people in England talk like the Queen...) On a side note, are there any rule sets for converting other variants of English (ie not USAlien:) phonemes to text? -- Atoms has finally been released! www.doldev.freeserve.co.uk-- Subcription/unsubscription/info requests: send e-mail with subject of "subscribe", "unsubscribe", or "info" to
Wear-Hard Mailing List Archive (searchable): http://wearables.blu.org
From Wear-Hard Mailing list Archive (WH)
Maintained by R. Paul McCarty
Archive created with babymail