Festival

The printable version is no longer supported and may have rendering errors. Please update your browser bookmarks and please use the default browser print function instead.

Festival is a general multi-lingual speech synthesis system developed at CSTR (Centre for Speech Technology Research). It 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, Italian, Czech and Spanish, with other languages available in prototype.)

Installation

Install the festival package. Voices are available with festival-us and festival-english packages.

German IMS

The IMS of the University Stuttgart developed an extension to Festival especially for German language (see here). It uses German voices with mbrola^AUR. To use it, install festival-ims^AUR with IMS Stuttgart patches. This should add support for the german voices de1 through de4. Install at least one of the voices (listed as optional dependencies), e.g. mbrola-voices-de2^AUR.

Configuration

There is no global /etc/ configuration file, but you can configure festival with your ~/.festivalrc file, or by directly editing /usr/share/festival/festival.scm. Both of these are scheme files, using scheme syntax and rerun everytime festival is run.

Sound server

The following allows Festival to work if audio from other sources is already playing. Add to your config:

For PulseAudio:

(Parameter.set 'Audio_Required_Format 'aiff)
(Parameter.set 'Audio_Method 'Audio_Command)
(Parameter.set 'Audio_Command "paplay $FILE --client-name=Festival --stream-name=Speech")

For ALSA: [1]

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

Voices

Arch splits the set of official voices into festival-us (recommended) and festival-english. The AUR has some others, in various states of maintenance which may or may not be currently working.

To see what voices are currently installed and what the default is, first enter Festival's #Interactive shell (a REPL scheme). To permanently change the default voice add it to your config, for example:

(set! voice_default voice_cmu_us_rms_cg)

Append to /usr/share/festival/siteinit.scm to apply for all users.

Manually

You can also get voices straight from Festvox [2] [3]. You will need to unzip and move the folder containing the voice to /usr/share/festival/voices/ and the way to tell what folder contains the voice is to look for a festvox/ subfolder inside of it. You can then test that your new voices are found by loading up the festival prompt.

Usage

To read a text file:

$ festival --tts text_file

To read a selection you highlighted with the cursor:

$ xsel | festival --tts

Convert a text file to an mp3 audio:

$ text2wave text_file | lame - text.mp3

Record audio with a select voice:

$ text2wave -o output.wav -eval '(voice_german_de2_os)' text_file

Interactive

Festival has an interactive prompt you can use for testing. Type festival to enter it. The following are some examples:

To show the voice festival speaks with:

voice_default

To list available voices:

(voice.list)

To select another voice, enter (voice_name). For example:

(voice_cmu_us_rms_cg)

To hear it speak:

(SayText "Arch makes me happy")

To list available commands:

help

To exit the shell:

(quit)

Troubleshooting

Can't open /dev/dsp

If festival returns the following error message:

Linux: can't open /dev/dsp

See #Sound server above.

Alsa playing at wrong speed

If the solution above gives you a squeaky voice, you might want to try changing your aplay options:

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

Command aplay not found

Install the alsa-utils package.

Killing festival process doesn't stop audio

Killing a background festival process will not stop the audio from continuing to play.

To stop audio from playing, the child audsp processes must be killed. This can be done by executing:

$ pkill -v audsp

Server

Install festival-freebsoft-utils^AUR to use Festival with speech-dispatcher (i.e. in Firefox's Reader).