Arch Linux

jwhendy

Member

Registered: 2010-04-01

Posts: 621

festival configuration help

Hi,

Just checking out festival. The WIKI mentions a "festivalrc" file, but I can't find any reference of where I might obtain a skeleton to start from, what options can go in one, or the like. "festival --help", for example, says nothing about how to set the voice. I know you can do it in the interactive mode, but I also don't see any mention of outputting to a file in the interactive mode when typing in "help."

I'm assuming that THIS is the official manual (or is it THIS?), it doesn't mention a config file or its options, either.

Lastly, flite is supposed to be the "lite" version... but it seems to have more options than festival, at least when looking at the output of "flite --help".

If anyone has experience... should I just stick with flite? Festival seems to have more voices, but flite just seems to be in better shape option-wise.

I guess I'm mainly looking for:
- where is the festivalrc supposed to go and what are it's options?
- does festival have the frequency, rate (words per minute), frequency, etc. options that flite has? Where are these?

Thoughts?

whitethorn

Member

Registered: 2010-05-02

Posts: 153

Re: festival configuration help

Hi,

I spent an evening playing around with festival.  It's been a couple months though, so I can only offer what I have in my .festivalrc.  Firstly the festivalrc should be in your users home folder.

namo ~/.festivalrc

I remember installing a couple voices from AUR, and her the contents of my festivalrc.  I remember finding a couple guides on how to use festival and the archwiki page

https://wiki.archlinux.org/index.php/Festival

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

jwhendy

Member

Registered: 2010-04-01

Posts: 621

Re: festival configuration help

Well, that helps with the location of festivalrc, and yes, those options are shown in the wiki. I guess it just felt like not much was presented in the way of documentation. Again, I was also surprised that flite had such a wide variety of options compared to festival. Note the difference:

]$ festival --help
Usage: festival Usage:
festival  <options> <file0> <file1> ...
In evaluation mode "filenames" starting with ( are evaluated inline
Festival Speech Synthesis System: 2.1:release November 2010
-q            Load no default setup files
--libdir <string>
              Set library directory pathname
-b            Run in batch mode (no interaction)
--batch       Run in batch mode (no interaction)
--tts         Synthesize text in files as speech
              no files means read from stdin
              (implies no interaction by default)
-i            Run in interactive mode (default)
--interactive
              Run in interactive mode (default)
--pipe        Run in pipe mode, reading commands from
              stdin, but no prompt or return values
              are printed (default if stdin not a tty)
--language <string>
              Run in named language, default is
              english, spanish and welsh are available
--server      Run in server mode waiting for clients
              of server_port (1314)
--script <ifile>
              Used in #! scripts, runs in batch mode on
              file and passes all other args to Scheme
--heap <int> {1000000}
              Set size of Lisp heap, should not normally need
              to be changed from its default
-v            Display version number and exit
--version     Display version number and exit

$ flite --help
flite: a small simple speech synthesizer
  Carnegie Mellon University, Copyright (c) 1999-2009, all rights reserved
  version: flite-1.4-release December 2009 (http://cmuflite.org)
usage: flite TEXT/FILE [WAVEFILE]
  Converts text in TEXTFILE to a waveform in WAVEFILE
  If text contains a space the it is treated as a literal
  textstring and spoken, and not as a file name
  if WAVEFILE is unspecified or "play" the result is
  played on the current systems audio device.  If WAVEFILE
  is "none" the waveform is discarded (good for benchmarking)
  Other options must appear before these options
  --version   Output flite version number
  --help      Output usage string
  -o WAVEFILE Explicitly set output filename
  -f TEXTFILE Explicitly set input filename
  -t TEXT     Explicitly set input textstring
  -p PHONES   Explicitly set input textstring and synthesize as phones
  --set F=V   Set feature (guesses type)
  -s F=V      Set feature (guesses type)
  --seti F=V  Set int feature
  --setf F=V  Set float feature
  --sets F=V  Set string feature
  -ssml       Read input text/file in ssml mode
  -b          Benchmark mode
  -l          Loop endlessly
  -voice NAME Use voice NAME
  -lv         List voices available
  -add_lex FILENAME add lex addenda from FILENAME
  -pw         Print words
  -ps         Print segments
  -pr RelName  Print relation RelName
  -v          Verbose mode