Re: IPA speech synthesizer

On Thu, Feb 19, 2009 at 10:41:24AM -0800, Sai Emrys wrote:
> I was looking at the Teonaht wikipedia article recently and thought:
> couldn't we make that IPA link autogenerate a synthesized sample?
>
> Surely someone has made an IPA synthesizer by now?
>
> And: how kludgy would it be if we just took a bunch of IPA samples -
> e.g. from http://www.phonetics.ucla.edu/course/chapter1/chapter1.html
> - and mashed 'em back to back?
>
> It wouldn't cover tone and lengthening and so on well, sure, but it'd
> be at least a first pass approximation.
This is, unfortunately, orders of magnitude more difficult to do than speech
synthesis for a specific language. :-(

I could write a lengthy essay on why this is impossible (or at least a Herculean
task), but the short version is as follows: the individual segments in a speech
stream affect each other so much that you can't just splice together phones and
get a result that sounds like speech.

--
Arnt Richard Johansen                                http://arj.nvg.org/
Someone just called to say he loved you?!

Replies