You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
asterisk/contrib
Russell Bryant b6bb13b498
Add gsm-devel as a package to install on redhat based systems.
14 years ago
..
editors add speex 16khz sample frame so codec cost can be calculated 14 years ago
init.d Merged revisions 298827 via svnmerge from 14 years ago
realtime Merged revisions 286588 via svnmerge from 14 years ago
scripts Add gsm-devel as a package to install on redhat based systems. 14 years ago
thirdparty
upstart Change all refererences to 1.6.3 to be 1.8, since that will be the next feature release 15 years ago
utils clean up a bunch more Zaptel-related references 16 years ago
README.festival
asterisk-doxygen-header
asterisk-ices.xml
asterisk-ng-doxygen Add an "Asterisk Architecture Overview" section to the doxygen documentation. 15 years ago
dictionary.digium
festival-1.4.1-diff
festival-1.4.2.diff
festival-1.4.3.diff
festival-1.95.diff
i18n.testsuite.conf Merged revisions 170671 via svnmerge from 16 years ago
valgrind-RedHat-8.0.supp
valgrind.supp Line 24 missed in compatibility fix in revision 233577 15 years ago

README.festival


app_festival is an application that allows one to send text-to-speech commands
to a background festival server, and to obtain the resulting waveform which
gets sent down to the respective channel. app_festival also employs a waveform
cache, so invariant text-to-speech strings ("Please press 1 for instructions")
do not need to be dynamically generated all the time.

You need :

1) festival, patched to produce 8khz waveforms on output. Patch for Festival
1.4.2 RELEASE are included. The patch adds a new command to festival
(asterisk_tts).

It is possible to run Festival without patches in the source-code. Just
add this to your /etc/festival.scm or /usr/share/festival/festival/scm:

(define (tts_textasterisk string mode)
"(tts_textasterisk STRING MODE)
Apply tts to STRING. This function is specifically designed for
use in server mode so a single function call may synthesize the string.
This function name may be added to the server safe functions."
(let ((wholeutt (utt.synth (eval (list 'Utterance 'Text string)))))
(utt.wave.resample wholeutt 8000)
(utt.wave.rescale wholeutt 5)
(utt.send.wave.client wholeutt)))

[See the comment with subject "Using Debian
festival >= 1.4.3-15 (no recompiling needed!)" on
http://www.voip-info.org/wiki-Asterisk+festival+installation for the
original mentioning of it]

2) You may wish to obtain and install the asterisk-perl
module by James Golovich <james@gnuinter.net>, from
either CPAN, or his site: http://asterisk.gnuinter.net,
as this contains a good example of how variable text
can be tts'd via asterisk, namely the examples/tts-*.agi
files there. It has been noted that the current expression
evaluation capabilities of asterisk are not best suited
for the generation and manipulation of text. AGI scripting
can be ideal for these sorts of needs. For simpler usage,
fixed, pre-recorded messages may be more amenable for your
purposes.

3) Before running asterisk, you have to run festival-server with a command
like :

/usr/local/festival/bin/festival --server > /dev/null 2>&1 &