Re: [orca-list] Orca: Support for Multilingual text
- From: Tomas Cerha <cerha brailcom org>
- To: orca-list gnome org
- Subject: Re: [orca-list] Orca: Support for Multilingual text
- Date: Mon, 19 Jul 2010 21:09:46 +0200
Bill Cox wrote:
On Mon, Jul 19, 2010 at 10:07 AM, Trevor Saunders
<trev saunders gmail com> wrote:
As has been discused previously I think generalized voice profiles
are better than language profiles, the only real difference being that
you could have different profiles for the same language.
Sounds good to me.
Yes, that's a logical extension of the same idea.
As for auto detection I'd be willing to consider choosing based on
char set, but I don't much like huristics beyond that.
I believe that's what voiceman does, and users seem to really like it,
so I'd say this would be a worthwhile enhancement.
It is a worthwhile enhancement, however I don't think it is the primary problem. I've
seen many discussions on this topic ending up concentrating just on this issue. What we
need first is the support for switching the languages. Whether we later develop
techniques to control it automatically is the next step, so let's not get stuck on it again.
Why don't we
just let the user tell us what language they'd like us to change to?
I think that voiceman lets you manually switch between Russian and
similar languages, and that determines how the Russian characters are
read. When English shows up, it's automatically read in English.
While this seems like a hack, it sounds like it serves the purposes of
many users quite well.
The success of autodetection very much depends on the set of languages you use. It may
work if you limit the possible languages to a few, which differ sufficiently as English
and Russian but would fail on many pieces of text for say English and French. It must
definitely be limited to the set of language profiles configured by the user and applied
sensibly. As it doesn't work for many combinations of languages, other means of
switching must be available anyway - manual switching and switching according to the
information provided by the text source (such as lang attribute in HTML, language
information in office documents, etc, etc.).
Best regards, Tomas
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]