Re: [g-a-devel] [Translation-i18n] Speech synthesis and i18n

[Samuel Thibault <samuel thibault ens-lyon org>]

Chusslove Illich, le Sun 13 Apr 2008 14:32:44 +0200, a écrit :
> > [...] some words, although written the same throughout the world (GNU,
> > Linux, Ubuntu, ...) [...]
> 
> This is not entirely valid assumption. In Serbian we would (well, we used
> to) transcribe it in writing (ГНУ, Линукс, Убунту, ...), removing any
> ambiguity in pronunciation on the voice level,

You would for text you write yourself, but not for web pages for
instance (unless there is automatic transcription).

> > [...] Speech synthesis, however, always use the native language
> > pronunciation, which results to the story above. [...]
> >
> >   #. Translators: this is the spoken word for Ubuntu, i.e. something that
> >   #. will be spoken the way Ubuntu would be pronounced in your language.
> >   msgid "Ubuntu"
> >   msgstr "Oubounetou"
> 
> In which PO would this fit? What you are describing to me seems as a
> "transcription dictionary" (something popular in my language, for the resons
> above).

It looks like indeed, but the difference is that the transcripted form
is _not_ to be used on a screen, but only for speech synthesis.

> It should probably be a standalone *something* for a given language,
> that people and machines could reference. (E.g. convertible into Orca's or
> other readers' pronunciation formats...)

Agreed.

Samuel