[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Emacs and Festival



For anyone who wants to check   for himself or herself, it is rather
easy with speechd-el, at least of you are on a Debian based
distribution. Just install the packages for festival, flite,
speech-dispatcher and speechd-el. To run emacs with speechd-el, I have
the following in my .emacs file:

;; speech dispatcher
(when (and (not (featurep 'emacspeak)) (getenv "SPEECHD"))
  (speechd-speak))

One can then start emacs with speechd-el with:

SPEECHD=1 emacs

One can switch between flite and festival simply by typing:

C-e d o module-name

Where module-name is either "flite" or "festival". I guess one could
also test espeak that way, or anything that is supported by speech-dispatcher.

On my 1Ghz machine wiht 512MB Ram, I still find festival a bit too
sluggish for efficient work, but I guess it is usable if no
alternative is available. I hope espeak will become a viable
alternative in the future especially for people who require languages
other than English.

Best regards, Lukas

Milan Zamazal writes ("Re: Emacs and Festival"):
> Although Festival may have more hardware requirements than other TTS
> systems, it's in my experience perfectly usable unless your computer is
> really old or it doesn't have enough RAM to store voices you use
> (together with applications you run, of course).  IIRC, my six years old
> PC with 256 MB RAM ran Festival with two languages, Emacs, X, etc. fine.
> 
> Nevertheless, applications shouldn't assume a TTS can perform just
> anything they decide to ask for.  Common mistake is feeding the Festival
> SayText function with a long text and waiting for the resulting sound.
> Festival should be asked to synthesize the text in chunks instead
> (festival-freebsoft-utils provides means for doing this easily) as it
> does in the tts_file function.
> 
> The only real performance problems with Festival I've ever met are:
> 
> - Festival used to be slow when echoing typed characters.  This can be
>   solved easily by caching the synthesized characters (this is what
>   Speech Dispatcher does).
> 
> - The default utterance chunking mechanism in Festival doesn't cope well
>   with wild input, such as a long text without any punctuation
>   characters or a complicated short text containing a lot of "strange"
>   characters.  The utterance chunking frontend in
>   festival-freebsoft-utils deals with most such situations (again, this
>   is what Speech Dispatcher uses).
> 
> So these problems are easily solvable and nobody needs to be afraid of
> Festival. :-)  Actually Festival with its high quality output, nice
> features, extensibility, several free voices and support of several
> languages should satisfy needs of many users.
> 
> Regards,
> 
> Milan Zamazal
> 
> -----------------------------------------------------------------------------
> To unsubscribe from the emacspeak list or change your address on the
> emacspeak list send mail to "emacspeak-request@cs.vassar.edu" with a
> subject of "unsubscribe" or "help"
> 

-----------------------------------------------------------------------------
To unsubscribe from the emacspeak list or change your address on the
emacspeak list send mail to "emacspeak-request@cs.vassar.edu" with a
subject of "unsubscribe" or "help"



If you have questions about this archive or had problems using it, please send mail to:

priestdo@cs.vassar.edu No Soliciting!

Emacspeak List Archive | 2007 | 2006 | 2005 | 2004 | 2003 | 2002 | 2001 | 2000 | 1999 | 1998 | Pre 1998

Emacspeak Files | Emacspeak Blog | Search the archive