Difference between revisions of "Festival"

Revision as of 23:57, 15 January 2013

Festival is a general multi-lingual speech synthesis system developed at CSTR (Centre for Speech Technology Research).

Festival offers a general framework for building speech synthesis systems as well as including examples of various modules. As a whole it offers full text to speech through a number APIs: from shell level, though a Scheme command interpreter, as a C++ library, from Java, and an Emacs interface. Festival is multi-lingual (currently British English, American English, and Spanish.)

Installation

Festival is available in the [extra] repository:

# pacman -S festival festival-english

Test festival:

$ echo "This is an example. Arch is the best." | festival --tts

If your hear all the example text, you sucessfully installed a TTS system.

If you do not hear anything, hear something strange, or get a message about /dev/dsp, see the Troubleshooting section.

Install another voice package

The installation described above will install the British English voice and set it as the default. To change it to something else, you must install another voice package.

American English voice:

# pacman -S festival-us

British English voice :

# pacman -S festival-english

Other voices are available for Festival; some of them can be found in the AUR.

Using with PulseAudio

Add these lines to the end of your .festivalrc file, or to /usr/share/festival/festival.scm :

(Parameter.set 'Audio_Required_Format 'aiff)
(Parameter.set 'Audio_Method 'Audio_Command)
(Parameter.set 'Audio_Command "pacat --channels=1 --rate=$SR $FILE -n Festival --stream-name Speech")

Configuration

You can set a default voice by adding these lines to the end of your ~/.festivalrc file, or to /usr/share/festival/festival.scm:

(set! voice_default 'voice_cmu_us_slt_arctic_hts)

You can also set the default voice in /usr/share/festival/voices.scm You must be root and the voices are at the end of the file. Just switch them around

Usage

Read a text file:

festival --tts /path/to/letter.txt

Read a text file to wav:

cat letter.txt | text2wave -o letter.wav

Interactive mode (testing voices etc.)

festival has an interactive prompt you can use for testing. Some examples (with sample output)

$ festival 
[...]
festival> 

List available voices:

festival> (voice.list)
(cstr_us_awb_arctic_multisyn kal_diphone don_diphone)

Set voice:

festival> (voice_cstr_us_awb_arctic_multisyn)
#<voice 0x1545b90>

Speak:

festival> (SayText '"test this is a test oh no a test bla test")
inserting pause after: t.
Inserting pause
[...]
id _63 ; name t ; 
id _65 ; name # ; 
#<Utterance 0x7f7c0c144810>

More:

festival> help 
"The Festival Speech Synthesizer System: Help

Quit: ctrl+d or

festival> (quit)

Example Script: Ping

One classic app that can make use of this is ping. Use this script to constantly ping a host, and return ping if success, fail is not:

#!/bin/bash
while [ 1 = 1 ]; do
     ping -c 1 $1 && (echo "Ping" | festival --tts) || (echo "Fail" | festival --tts)
done

Note that this does not really work on multisynth voices, as they take a while to prepare before playing.

Troubleshooting

Can't open /dev/dsp

If festival returns the following error message:

Linux: can't open /dev/dsp

Switch to ALSA output by adding these lines to the end of your .festivalrc file, or to /usr/share/festival/festival.scm (source):

(Parameter.set 'Audio_Method 'Audio_Command)
(Parameter.set 'Audio_Command "aplay -q -c 1 -t raw -f s16 -r $SR $FILE")

Alsa playing at wrong speed

If the solution above gives you Mr. Squeaky Voice, you might want to try:

(Parameter.set 'Audio_Method 'Audio_Command)
(Parameter.set 'Audio_Command "aplay -Dplug:default -f S16_LE -r $SR $FILE")

External links

• Festival manual.