Fred Roller froller at
Tue Jun 9 21:41:31 UTC 2009

On Tue, 2009-06-09 at 14:27 -0600, CJ Kelley wrote:
> On Tue, Jun 9, 2009 at 2:13 PM, Tom Rausner <tom at> wrote:
>         tir, 09 06 2009 kl. 21:05 +0100, skrev Muzer:
>         > Just use festival! It is SO much better!
>         year, I think I've reached that piont too...
>         --
>         Tom Rausner <tom at>
>         --
>         ubuntu-users mailing list
>         ubuntu-users at
>         Modify settings or unsubscribe at:
> ok so when i type 
> festival --tts test.txt
> it tells me Linux: Can't open /dev/dsp
> what process is using it and how can i stop it?
> --cj
> -- 
> Registered linux user #402,916 (registered since 2007)

How is festival for the "Human" factor.  I had tried it before but was
JUST learning linux.  Now I am on a quest to convert pdf's to mp3, busy
life.  Currently, as mentioned in an earlier post I use espeak from the
repository and mbrola (which helps quite a bit in adding the human
feel.)  But with espeak I still get a decent voice with

espeak -a 120 -p 30 -s 190 -k 15 -g 12 -v english -m -w /tmp/tmp.wav -f
test_prose |mplayer /tmp/tmp.wav

options are: (espeak -h)
-a = amplitude
-p = pitch
-s = speed in words per minute
-k = adjust pitch for Capital letters
-g = word gap
-v = specific voice from list
-m = ignore hypertext tags
-w = create .wav file
-f = use this file for input

This should get you the package.
sudo apt-get install espeak

mbrola (from: ) did well on
creating a softer easier to listen to voice.  There website took a bit
to get around but once I found got the software and the "voice" it was
fairly easy to work with and use.  Espeak has voices which pair with
mbrola "dictionary" db.  If you would like a sample go to . I have put a chapter from linux study guide
there for you to try.

Fred R.

"Life is like Linux, simple.  If you are fighting it, you are doing
something wrong."

More information about the ubuntu-users mailing list