Re: VoiceInputStream proposal

["Marisha Ray & Neil Jerram" <mpriz dircon co uk>]

Vazquez,

This sounds like a very interesting idea!  A few comments...

1.  Have you seen EmacsSpeak by T.V.Raman?  I believe that this can speak
most of the output that Emacs produces, so it is almost a complete solution
for people who do everything in Emacs.  Being Emacs-based, it also uses the
major and minor modes as contextual information to modify how text is
spoken.

2.  Speech technology is improving very quickly, so I strongly recommended a
"pluggable" architecture, where different "back-end" speech engines can be
swapped with each other without having to change the API that
applications/Gnome uses to produce speech.

3.  At what level in the X / GTK / Application hierarchy should you try to
integrate speech output?  In my experience, useful speech is very
context-dependent.  It might be helpful to speech-enable some GTK widgets -
e.g. widgets with a current "selection" could speak the new selection
whenever the selection changes - but I think that most speech output will
need to be controlled by the application.

Regards,

    Neil

-----Original Message-----
From: Vazquez Gonzalez <lsc98025@lab.dit.upm.es>
To: gnome-list@gnome.org <gnome-list@gnome.org>
Date: 12 November 1998 14:21
Subject: VoiceInputStream proposal


>View attach.
>
>jose
>