VoiceText Support¶
Cepstral¶
Also known as Text2Speech, DomotiGa currently supports 2 speech engines, the default 'espeak' festival command, standard installed on Ubuntu.
And the commercial Cepstral engine called 'swift'.
Install your swift engine and voices like this:
$ mkdir -p ~/install/cepstral
$ cd ~/install/ceptral
$ wget http://downloads.cepstral.com/cepstral/i386-linux/Cepstral_Linda_i386-linux_5.1.0.tar.gz
$ tar xzvf Cepstral_Linda_i386-linux_5.1.0.tar.gz
$ cd Cepstral_Linda_i386-linux_5.1.0
$ sudo ./install.sh
Install into what directory? [/opt/swift":http://www.cepstral.com/|here]] Enter
/opt/swift does not exist. Create it? ([n]/y) y
Swift will be installed in the following directories:
Voices in /opt/swift/voices
Shared libraries in /opt/swift/lib
Binaries in /opt/swift/bin
Configuration file in /opt/swift/etc
Header files in /opt/swift/include
Examples in /opt/swift/examples
Sound effects filters in /opt/swift/sfx
Documentation in /opt/swift/doc
Is this acceptable? Enter 'yes' to continue: yes
Installing libraries...
***************************************************************************
If you are installing Swift system-wide, you may need to add the following
line to /etc/ld.so.conf and run ldconfig as root:
/opt/swift/lib
(Otherwise, you will need to add it to the LD_LIBRARY_PATH environment
variable in order to run programs linked against the Swift libraries.)
***************************************************************************
Installing voice Linda...
Creating configuration...
Installing binaries...
Installing symbolic link to swift...
Installing man page...
Setting permissions...
Kill License Server
Testing the installed swift binary...
/opt/swift/bin/swift -o /dev/null 'hello world'
***************************************************************************
****************** Installation Completed Successfully! *******************
***************************************************************************
To hear a test:
$ /opt/swift/bin/swift 'hello world'
If you have licenses, register them with this command:
$ sudo swift --reg-voice
Repeat above steps for other Cepstral voices you have.
I have bought voices Emily and Linda, they sound great!
Enable support via the VoiceText Setup menu, choose which engine and which voice(s) you want to use.
Test them with the buttons on the System Debug page.
espeak¶
picotts¶
Picotts is a small lightweight TTS engine ported from Android.
On Ubuntu you can install it like this:
$ sudo apt-get install libttspico-utils
In DomotiGa's voicetext settings please specify any of the supported languages in the Lang field.
$ pico2wave -l en-US -w /tmp/test.wav "Just a test!"
Valid languages:
en-US
en-GB
de-DE
es-ES
fr-FR
it-IT
Usage: pico2wave <words>
-w, --wave=filename Write output to this WAV file
-l, --lang=lang Language (default: "en-US")
Help options:
-?, --help Show this help message
--usage Display brief usage message
Thanks to MrGoodFriend for adding/sharing picotts support!
Related Resources¶
Comment¶
Updated by: rdnzl, Updated 3 months ago
Access count: 642 since 2011-08-25
