Re: IPA speech synthesizer

----- Original Message ----
> From: Philip Newton <philip.newton@...>
> So that, for example, "conlang" would be, conceptually, produced by
> splicing together the bigrams for [kQ] [Qn] [nl] [l&] [&N]. (Not sure
> whether [#k] and [N#] would also get bigrams; possibly so.)
>
> This would go some way toward taking care of the "troughs" that BPJ
> referred to, since you don't have, say, [&] spoken in isolation but as
> part of the bigrams [l&] and [&N], which take care of the transitions.
Wouldn't that potentially shift the gaps, though, rather than getting rid of them? At
any rate, it's potentially deceptive, since both of the vowels in "conlang" are
at least partially nasalized due to the following nasal consonants (cf. "con"
vs. "cod"). For con, cod, and cot, we would do one of these:

[kQ][Qt], [kQ][Q:d], [kQ][Q~n], where the computer would then blend the Q and Q~ somehow
[kQ][Qt], [kQ:][Q:d], [kQ~][Q~n], which has the advantage of requiring less reliance on
the computer but generates a lot more needed bigrams, many of them rarely if
ever used

Ultimately, it seems like "cop," "cob," "com," "con," "cot," "cod," "cog," and "cock"
represent 8 subtly different manifestations of /o/ (and that's without changing
the initial consonant :) ... "bop" vs "cop" vs "mop" is a further
complication).

In practice, as far as the actual sound wave is concerned, it seems to me that
stops (in particular) are more distinguished from each other in their effect
on the surrounding sounds than they are in their own sound (i.e., lack
thereof).

-- Paul