Speech Synthesis

Sannyasin Brahmanathaswami brahma at hindu.org
Sun Apr 5 15:43:34 EDT 2020


Given the arrival of the Speech Library.... Something I played with 30 years ago on hypercard.... manually entering phonemes,  to get the computer talk was a fascinating thing. Today, with AI all around us, Speech Sythesis is "everywhere"  and very advanced,

When we initialize

speechLibraryInitialized pSuccess

Do we have gain control of the International Phonetic Alphabet?  English is a "creole/polyglot" language, and depending across what language/culture "adjacencies"  (English-Tamil, English-French, English-Spanish) that the discourse it in, X number of words come into play that are understand, and pronounced correctly, but both parties. And the polyglot is increasing monthly! I don't know how AI will ever keep up.

Above is background. 

What I want is:  to have "Fred"  (or "Junior" or "Deranaged") pronounce "Satchinananda" perfectly.

Or put another way: having access to

com.apple.speech.synthesis.voice.Deranged

Does Deranged have command over any combination of phonemes?

I can't imagine we have to start from ground zero on this. Has anyone else done work in this area.

BR




More information about the use-livecode mailing list