[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Enabling Outloud on Fedora 20

Dear Friends:

This email is to document the steps I took to enable Emacspeak to use
ViaVoice on my 64-bit Fedora 20 (beta) laptop. ViaVoice is separately
available as a proprietary TTS product called Voxin today, and has
previously been sold under the name TTSynth. It has also been known by
the name Eloquence on certain operating systems.

I document these steps here because several script edits were needed.
Perhaps this can be obviated in the future by introducing conditional
statements in these scripts?  Also, it was necessary to manually acquire
and copy the 32bit TCL executable saving it to a filename the script
expected. This last action is probably the most arcane step I had to
take. Perhaps Fedora should be encouraged to include the 32bit TCL
binary in its 64bit package--because simply installing the 32bit TCL
rpm is insufficient!

These steps do not include acquisition, installation and proper
configuration of ViaVoice. Rather, it is presumed this TTS engine has
been correctly installed and is known to be functioning.

NOTE: Once installed, it is possible to use this
TTS engine with Emacspeak in one console session while also using Speakup, through the
Speakup Connector, in any number of other console sessions without
unmanageable conflict. I have NOT, however, been successful in using this engine with Emacspeak (or Speakup) and Orca
via Speech-Dispatcher.

I note that I am using this TTS engine with alsa, not with pulseaudio. I
have no knowledge of whether these steps will allow this TTS engine to
be used with pulseaudio. On my systems pulseaudio is always terminated
with extreme prejudice, i.e.

rm -f /usr/bin/pulseaudio
touch /usr/bin/pulseaudio
chmod 444 /usr/bin/pulseaudio

Lastly, please note that all commands provided in this message are assumed executed as root, including those immediately preceding.

My thanks to T.V. Raman, Jason White, and Bill Acker for helping
me discover and implement all the relevant details required to make
Emacspeak work with ViaVoice over alsa on Fedora 20.


Step 1:	Install the 32bit tclx libraries.

yum install tclx.i686

Step 2:	Acquire and Install the 32bit TCL executable

In a temporary directory somewhere:

mkdir tmp
cd tmp

	a.)	Acquire the 32bit rpm package

	yumdownloader tcl.i686

	b.)	Use cpio2rpm to populate a file hierarchy with all files
	from the tcl.i686 rpm. In the command below, [filename.rpm] is
	the tcl.i686 rpm just downloaded in substep a.) above:

rpm2cpio [filename.rpm] |cpio -imd | 

	c.)	Identify the 32bit TCL binary and install it to an
	appropriate system directory. We are not surprised to find this
	executable in usr/bin/ following substep b.) above. We place
	this file in /usr/local/bin/ because it's the identified locus
	of user installed and managed binaries. We also rename this file
	as we copy it, because our Emacspeak speech server script,
	/usr/share/emacs/site-lisp/emacspeak/32-outloud expects to find
	it in /usr/bin/ and expects it to be called tclsh8.4_ia32--even
	though it's from a TCL version 8.5:

	cp -a usr/bin/tclsh8.5 /usr/bin/tclsh8.4_ia32

NOTE: Do not make the mistake of trying to copy the symbolic link called
tclsh. Also beware that a TCL version value is encoded in the
executable's filename, so the exact filename you will copy may vary.

	d.)	Insure your copied executable will be executable by all:

	chown root:root /usr/bin/tclsh8.4_ia32
	chmod 755 /usr/bin/tclsh8.4_ia32

	e.)	It is now safe to remove the temporary directory, with
	all included files, you used in the substeps above:

	cd ..
	rm -rf tmp

Step 3:	Insure required files are present.

If you installed Emacspeak from the Fedora rpm, you already have
However, you do not have two essential script files in

Obtain the two script files 32-outloud and outloud from the Emacspeak
sbn repository, or from an Emacspeak tarball, and place in the servers

If you obtained Emacspeak from sbn (preferred by long-time Emacspeak
users), or from the Emacspeak tarball, you will find you already have
these files.

Step 4: Edit your Emacspeak speech server script

cd /usr/share/emacs/site-lisp/emacspeak/servers

	a.)	Edit line 45 of 32-outloud to point to the .so file
	where Fedora locates it. As of Fedora 20 this should read:

	load /usr/lib/tcl8.5/tclx8.4/libtclx8.4.so

		b.)	Edit line 49 of 32-outloud to reference the
		32bit Fedora TCL package name. As of Fedora 20 this is:

		package require Tclx


Janina Sajka,	Phone:	+1.443.300.2200
		Email:	janina@rednote.net

Linux Foundation Fellow
Executive Chair, Accessibility Workgroup:	http://a11y.org

The World Wide Web Consortium (W3C), Web Accessibility Initiative (WAI)
Chair,	Protocols & Formats	http://www.w3.org/wai/pf
	Indie UI			http://www.w3.org/WAI/IndieUI/

To unsubscribe from the emacspeak list or change your address on the
emacspeak list send mail to "emacspeak-request@cs.vassar.edu" with a
subject of "unsubscribe" or "help".

If you have questions about this archive or had problems using it, please send mail to:

priestdo@cs.vassar.edu No Soliciting!

Emacspeak List Archive | 2010 | 2009 | 2008 | 2007 | 2006 | 2005 | 2004 | 2003 | 2002 | 2001 | 2000 | 1999 | 1998 | Pre 1998

Emacspeak Files | Emacspeak Blog