XVoice

Version 0.8, Thu Mar  9 14:17:37 PST 2000
by Brian Craft, while Tom is tied up on other things.
http://www.thecraftstudio.com/bcboy/xvoice
bcboy@thecraftstudio.com

Version 0.6, Nov 25, 1999.
http://www.compapp.dcu.ie/~tdoris/Xvoice
Tom Doris
tdoris@compapp.dcu.ie

Versions previous to 0.5 by:
http://www.zachary.com/creemer/xvoice.html
David Z. Creemer
david@zachary.com

XVoice accepts contuinuous speech input from IBM's ViaVoice runtime for Linux,
and then retarget the resulting text at many X applications. It uses the
XTest extension to accomplish this.

Speech input can be commands, such as "delete three lines", or dictation, such
as "You sent the attachment in Microsoft Word format, a secret proprietary
format, so I cannot read it." XVoice will map commands to key and mouse events,
as specified in the configuration file. It will map dictation to equivalent key
events, as if you had typed it. Command mode and dictation mode can be enabled
independently.

The ViaVoice engine requires some environment variables to be set. XVoice will
attempt to bootstrap the environment by invoking vvsetenv (shipped with the
ViaVoice run-time kit) and restarting. If this fails you need to set up your
environment by executing the following in your (bourne syntax) shell:

shell% . vvsetenv

XXX -- Is the following true? or just an XSendEvent relic?

In order to get emacs to work, you _must_ evaluate the following
elisp:

(setq x-allow-sendevents t)

BUILDING

XVoice requires X, ViaVoice SDK, GTK+, and libgnomeui to build. Some of the
code is based on the IBM ViaVoice SDK sample "gtkhello". If you can build that
sample program, you can build XVoice. See the Makefile for local settings.

RUNNING

"xvoice -h" will show you help on running the application.

"xvoice -m" starts XVoice with the microphone ON.

XVoice

xvoice actively tracks the focus window and loads whatever grammars it finds
for that window. It also keeps a history of any context grammars that were
loaded, so if the focus changes the context grammars can be automatically
reloaded when the window is re-focused. "command mode" and "dictate mode" may
be used to control which vocabularies are loaded.

Once started, XVoice presents a list of currently active vocabularies,
a list of recognised words (initially blank), and the current target
application (initially blank). Focus a target with the mouse, or by
saying "next window" until the desired application is focused, and say
"command mode" or "dictate mode" to begin sending voice commands.
Say "stop command" or "stop dictate" to stop each mode.

"Microphone off" will turn off the microphone.

While dictating into an application, speaking "stop dictation" will
return to XVoice command mode. Speaking "correction" will erase the
last dictated word or punctuation. You may erase words back to the
beginning of the dictation session. Speaking "new paragraph" will
enter a single CR for you (i.e. "enter").

Speaking "command mode" will put Xvoice into application command mode.  Xvoice
will attempt to find the grammar file for the focused window, and use this as
the current grammar. The xvoice.xml file contains mappings from spoken text to
synthesisted X events. Thus it is possible to speak "open file" and have the
key sequence "C-x C-f" sent to emacs. It is possible for users to modify the
xvoice.xml file to incorporate key sequences for other applications.  This
procedure is described in the COMMANDS file.

COPYRIGHT

Long Version:

XVoice is Copyright (c) 1999, 2000, 2001, 2002 by David Z. Creemer, Tom Doris,
Brian Craft, Jessica Heckman, Deborah Kaplan.

All rights not granted in the LICENSE file are reserved. XVoice is distributed
subject to the terms explained in the LICENSE file. Portions of this program
are based on code from other projects, which use the GPL and other licenses.
All attributions are in the AUTHORS file.

Short Version: GPL.