Re: [orca-list] Orca and character encoding

[Helios <heliooos gmail com>]

Hi Tomas,
I think it was Vinux development list, because I read this topis today :-)
the thread link is here:
http://groups.google.com/group/vinux-development/browse_thread/thread/6acd8527a3bf4fc5?hl=en

Currently I do not have possibility to test Debian based Vinux 2.0. In
older Ubuntu based Vinux 1.4 there was no such problem I think. But
because of problems with pulseaudio I used Espeak with Gnome speech
and not with SD.

Helios
------------------------------

Message: 3
Date: Fri, 18 Sep 2009 10:39:31 +0200
From: Tomas Cerha <cerha brailcom org>
To: Isaac Porat <isaac porat me uk>
Cc: Orca-list gnome org
Subject: Re: [orca-list] Orca and character encoding
Message-ID: <4AB34743 3090306 brailcom org>
Content-Type: text/plain; charset=UTF-8

Isaac Porat wrote:
> On another mailing list, there are reports on problems with utf-8 and it
> suggested that the problem is perhaps with speech-dispatcher but it is not
> clear.

Which list?

> I wonder how Orca handles character encoding, does Orca translate all
> characters to a common denominator such as utf-8 and sends these to the
> speech engine or does it send it in it in its original encoding and if so
> how it indicates the encoding to the engine?

Orca internally works with utf-8 encoded strings and this encoding is
also used for communication with Speech Dispatcher.  The Speech
Dispatcher Python interface accepts either utf-8 encoded strings or
Python unicode strings, but in either case, the strings are encoded to
utf-8 for communication with the SD server core internally.  SD then
recodes to the encoding accepted by the speech engine, which ofen is
utf-8 again, but may differ according to what the speech engine supports.

Best regards, Tomas



------------------------------