HOWTO: Make festival TTS use better voices (MBROLA / CMU / HTS)

[durand]

Thanks, this guide work perfectly on Ubuntu Lucid.

[hansum_rahul]

Thank you!

Just to mention: the /etc/festival.scm did not exist on my system, instead I modified this:

/usr/share/festival/festival.scm

On a Hardy

[socratesfoot]

I could not get the default voice to work in the festival.scm. To set the default voice I edited the init.scm.
If you're looking to automate the process of setting the default voice. I used the "sed" command.

sudo cp /usr/share/festival/init.scm /usr/share/festival/init.scm.backup
sudo sed 's/(eval(list voice_default))/(eval (list voice_nitech_us_slt_arctic_hts))/' init.scm > tmp_file
sudo mv tmp_file init.scm

[socratesfoot]

Sorry, to double post but I was thinking about it and it's probably better to just post the whole thing from my history, since the Festival configuration files are in a slightly different location with Ubuntu 10.

Note, I only wanted the "Female English" voice, so that's all this installed. These voice files are huge and I didn't see any reason to put them all on.

sudo apt-get install festlex-poslex festival festlex-cmu
sudo cd /usr/share/festival/voices/english
sudo mkdir hts_tmp
cd hts_tmp/
sudo wget -c http://hts.sp.nitech.ac.jp/archives/2.1/festvox_nitech_us_slt_arctic_hts-2.1.tar.bz2
sudo wget -c http://hts.sp.nitech.ac.jp/archives/1.1.1/cmu_us_kal_com_hts.tar.gz
sudo wget -c http://hts.sp.nitech.ac.jp/archives/1.1.1/cstr_us_ked_timit_hts.tar.gz
for t in `ls` ; do sudo tar xvf $t ; done
sudo rm festvox_nitech_us_slt_arctic_hts-2.1.tar.bz2 cmu_us_kal_com_hts.tar.gz cstr_us_ked_timit_hts.tar.gz
sudo mkdir -p /usr/share/festival/voices/us
sudo mv lib/voices/us/* /usr/share/festival/voices/us/
sudo mv lib/hts.scm /usr/share/festival/hts.scm
cd ..
sudo rm -rf hts_tmp/
sudo cp /usr/share/festival/init.scm /usr/share/festival/init.scm.backup
sudo sed 's/(eval(list voice_default))/(eval (list voice_nitech_us_slt_arctic_hts))/' init.scm > tmp_file
sudo mv tmp_file init.scm

[Hendrixski]

Tried the script... after fixing a few errors (e.g. there is no "sudo cd"... and you have to create tmp_file with adequate permissions before sudo sed can use it)

Whenever I try to play a sound I get this crap:

festival> (voice_cstr_us_ked_timit_hts)    
cstr_us_ked_timit_hts
festival> (SayText "hello, this is a test")
Segmentation fault
Cannot open file /tmp/est_18045_00001/utt.wav as tokenstream
Wave load: can't open file "/tmp/est_18045_00001/utt.wav"
Cannot load wavefile: /tmp/est_18045_00001/utt.wav
#<Utterance 0xb6b35868>
festival>

What's wrong?

Is there a PPA for these as a package? (because I haven't done so much work to install something since I used Windows.. .this is NOT the experience I expect from Ubuntu)

[VV Cephei]

Sweet. This is exactly what I was looking for. Thanks!

I am having a weird problem with festival, though. I haven't seen anything on the internet, so I thought I'd start here:

If I feed festival a file, either with
cat testfile | esddsp festival --tts
or by first opening festival and using:
(tts_file "testfile")

It seems to divide the file into a bunch of strings delimited by newlines or punctuation, and then says all the strings _at the same time_. However, if I use:
(SayText "reset
> john-laptop pulseaudio: ratelimit.c: events suppressed
> john-laptop kernel: lo: Disabled Privacy Extensions
> john-laptop kernel: lo: Disabled Privacy Extensions
> 08:33:20 AM
> john-laptop pulseaudio: ratelimit.c: events suppressed
> john-laptop pulseaudio: ratelimit.c: events suppressed")

then it works just fine. To be clear, the last command works perfectly, but the previous two result in festival saying everything in the file simultaneously. Anyone else have this problem?

[jambel]

Nitech HTS voices doesn't seem to work any more on Ubuntu 10.10 and festival 2.0.95 from ubuntu repos.

After these instructions:

mkdir hts_tmp
cd hts_tmp/
wget -c http://hts.sp.nitech.ac.jp/archives/...ts-2.1.tar.bz2
wget -c http://hts.sp.nitech.ac.jp/archives/...ts-2.1.tar.bz2
wget -c http://hts.sp.nitech.ac.jp/archives/...ts-2.1.tar.bz2
wget -c http://hts.sp.nitech.ac.jp/archives/...ts-2.1.tar.bz2
wget -c http://hts.sp.nitech.ac.jp/archives/...ts-2.1.tar.bz2
wget -c http://hts.sp.nitech.ac.jp/archives/...ts-2.1.tar.bz2
wget -c http://hts.sp.nitech.ac.jp/archives/...com_hts.tar.gz
wget -c http://hts.sp.nitech.ac.jp/archives/...mit_hts.tar.gz
Unpacking the voices

Next we'll unpack the voices:


Code:for t in `ls` ; do tar xvf $t ; done
Installing the voices

Now we can install the voices:

sudo mkdir -p /usr/share/festival/voices/us
sudo mv lib/voices/us/* /usr/share/festival/voices/us/
sudo mv lib/hts.scm /usr/share/festival/hts.scm

and set this voice (set! voice_default 'voice_nitech_us_slt_arctic_hts) as default on /etc/festival.scm I face that error...

Error: HTS_Model_load_pdf: Failed to load header of pdfs.

Any ideas to refresh the instructions of this topic?

Cheers,

John.

[digitaltoast]

http://hts.sp.nitech.ac.jp/archives/2.1/festvox_nitech_us_awb_arctic_hts-2.1.tar.bz2

Error: HTS_Model_load_pdf: Failed to load header of pdfs.

I had this same problem - until I found a post that suggested that it's because festival 2.095 requires HTS 2.1.1 voices, which can be found here:
http://hts.sp.nitech.ac.jp/archives/2.1.1/

But it's not straightforward! The whole Festival system seems to be designed to be complicated and keep non-geeks out!

Want to try the 2.1.1 voices? You need to do this:

* Installation of HTS-demo_CMU-ARCTIC-SLT
==========================================

1. HTS-demo_CMU-ARCTIC-SLT requires Festival, SPTK-3.3, HTS-2.1.1, hts_engine API-1.03, and OpenFst-1.1.
   Please install them before running this demo.
   You can download them from the following websites:

   Festival: http://www.cstr.ed.ac.uk/projects/festival/
   SPTK: http://sp-tk.sourceforge.net/
   HTS: http://hts.sp.nitech.ac.jp/
   hts_engine API: http://hts-engine.sourceforge.net/
   OpenFst: http://www.openfst.org/

   In HTS-demo_CMU-ARCTIC-SLT, a simple F0 extraction script written in Tcl/Tk is included.
   This script calls get_f0 function implemented in the open-source speech toolkit Snack.
   Therefore, HTS-demo_CMU-ARCTIC-SLT also requires Tcl/Tk with Snack.
   ActiveState (http://www.activestate.com/) provides a Tcl/Tk distribution named ActiveTcl
   for many platforms.  You can download it from

   ActiveTcl: http://downloads.activestate.com/ActiveTcl/

   The above distribution includes Snack and it is easy to install and use.
   We recommend you to use this to run this demonstration
   (Of course you can use your own tcl/tk with Snack).
   Note that ActiveTcl 8.5 doesn't include Snack, please use ActiveTcl 8.4.


2. Setup HTS-demo_CMU-ARCTIC-SLT by running configure script:

   % cd HTS-demo_CMU-ARCTIC-SLT
   % ./configure --with-tcl-search-path=/usr/local/ActiveTcl/bin \
                 --with-fest-search-path=/usr/local/festival/examples \
                 --with-sptk-search-path=/usr/local/SPTK-3.3/bin \
                 --with-hts-search-path=/usr/local/HTS-2.1.1_for_HTK-3.4.1/bin \
                 --with-hts-engine-search-path=/usr/local/hts_engine_API-1.03/bin \
                 --with-openfst-search-path=/usr/local/openfst-1.1/bin

   Please adjust the above directories for your environment.
   Note that you should specify festival/examples rather than festival/bin.

   You can change various parameters such as speech analysis conditions and model training conditions
   through ./configure arguments.  For example

   % ./configure MGCORDER=24 GAMMA=0 FREQWARP=0.0              (24-th order cepstrum)
   % ./configure MGCORDER=24 GAMMA=0 FREQWARP=0.42             (24-th order Mel-cepstrum)

   % ./configure MGCORDER=12 GAMMA=1 FREQWARP=0.0  LNGAIN=0    (12-th order LSP,     linear gain)
   % ./configure MGCORDER=12 GAMMA=1 FREQWARP=0.0  LNGAIN=1    (12-th order LSP,     log gain)
   % ./configure MGCORDER=12 GAMMA=1 FREQWARP=0.42 LNGAIN=1    (12-th order Mel-LSP, log gain)
   % ./configure MGCORDER=12 GAMMA=3 FREQWARP=0.42 LNGAIN=1    (12-th order MGC-LSP, log gain)

   % ./configure NSTATE=7 NITER=10 WFLOOR=5   (# of HMM states=7, # of EM iterations=10, mix weight floor=5)

   Please refer to the help message for details:

   % ./configure --help


3. Start running demonstration as follows:

   % cd HTS-demo_CMU-ARCTIC-SLT
   % make

   After composing training data, HMMs are estimated and speech waveforms are synthesized.
   It takes about 12 to 18 hours :-)

12 to 18 HOURS??? And I don't even know what I'm going to end up with. What does "DEMO" mean? Does it just say something and stop? Also, do I want
http://hts.sp.nitech.ac.jp/archives/...-ADAPT.tar.bz2
or
http://hts.sp.nitech.ac.jp/archives/...RAIGHT.tar.bz2
?

It's not the 492Mb of each file I mind, it's the idea of spending 12-18 hours building one to find I wanted the other one!

The only manual I can find for Festival is here:
http://www.cstr.ed.ac.uk/projects/festival/manual/
Dated 1999, for version 1.4

I sometimes feel like I've missed the basics somewhere.
Were it not for threads like this I'd be completely lost!

[jambel]

I still haven't make the nitech voices to work but I tried those and worked fine, just their not what I prefer.

try to follow the README file and if you have any issue, nudge me!

I had this same problem - until I found a post that suggested that it's because festival 2.095 requires HTS 2.1.1 voices, which can be found here:
http://hts.sp.nitech.ac.jp/archives/2.1.1/

But it's not straightforward! The whole Festival system seems to be designed to be complicated and keep non-geeks out!

Want to try the 2.1.1 voices? You need to do this:

* Installation of HTS-demo_CMU-ARCTIC-SLT
==========================================

1. HTS-demo_CMU-ARCTIC-SLT requires Festival, SPTK-3.3, HTS-2.1.1, hts_engine API-1.03, and OpenFst-1.1.
   Please install them before running this demo.
   You can download them from the following websites:

   Festival: http://www.cstr.ed.ac.uk/projects/festival/
   SPTK: http://sp-tk.sourceforge.net/
   HTS: http://hts.sp.nitech.ac.jp/
   hts_engine API: http://hts-engine.sourceforge.net/
   OpenFst: http://www.openfst.org/

   In HTS-demo_CMU-ARCTIC-SLT, a simple F0 extraction script written in Tcl/Tk is included.
   This script calls get_f0 function implemented in the open-source speech toolkit Snack.
   Therefore, HTS-demo_CMU-ARCTIC-SLT also requires Tcl/Tk with Snack.
   ActiveState (http://www.activestate.com/) provides a Tcl/Tk distribution named ActiveTcl
   for many platforms.  You can download it from

   ActiveTcl: http://downloads.activestate.com/ActiveTcl/

   The above distribution includes Snack and it is easy to install and use.
   We recommend you to use this to run this demonstration
   (Of course you can use your own tcl/tk with Snack).
   Note that ActiveTcl 8.5 doesn't include Snack, please use ActiveTcl 8.4.


2. Setup HTS-demo_CMU-ARCTIC-SLT by running configure script:

   % cd HTS-demo_CMU-ARCTIC-SLT
   % ./configure --with-tcl-search-path=/usr/local/ActiveTcl/bin \
                 --with-fest-search-path=/usr/local/festival/examples \
                 --with-sptk-search-path=/usr/local/SPTK-3.3/bin \
                 --with-hts-search-path=/usr/local/HTS-2.1.1_for_HTK-3.4.1/bin \
                 --with-hts-engine-search-path=/usr/local/hts_engine_API-1.03/bin \
                 --with-openfst-search-path=/usr/local/openfst-1.1/bin

   Please adjust the above directories for your environment.
   Note that you should specify festival/examples rather than festival/bin.

   You can change various parameters such as speech analysis conditions and model training conditions
   through ./configure arguments.  For example

   % ./configure MGCORDER=24 GAMMA=0 FREQWARP=0.0              (24-th order cepstrum)
   % ./configure MGCORDER=24 GAMMA=0 FREQWARP=0.42             (24-th order Mel-cepstrum)

   % ./configure MGCORDER=12 GAMMA=1 FREQWARP=0.0  LNGAIN=0    (12-th order LSP,     linear gain)
   % ./configure MGCORDER=12 GAMMA=1 FREQWARP=0.0  LNGAIN=1    (12-th order LSP,     log gain)
   % ./configure MGCORDER=12 GAMMA=1 FREQWARP=0.42 LNGAIN=1    (12-th order Mel-LSP, log gain)
   % ./configure MGCORDER=12 GAMMA=3 FREQWARP=0.42 LNGAIN=1    (12-th order MGC-LSP, log gain)

   % ./configure NSTATE=7 NITER=10 WFLOOR=5   (# of HMM states=7, # of EM iterations=10, mix weight floor=5)

   Please refer to the help message for details:

   % ./configure --help


3. Start running demonstration as follows:

   % cd HTS-demo_CMU-ARCTIC-SLT
   % make

   After composing training data, HMMs are estimated and speech waveforms are synthesized.
   It takes about 12 to 18 hours :-)

12 to 18 HOURS??? And I don't even know what I'm going to end up with. What does "DEMO" mean? Does it just say something and stop? Also, do I want
http://hts.sp.nitech.ac.jp/archives/...-ADAPT.tar.bz2
or
http://hts.sp.nitech.ac.jp/archives/...RAIGHT.tar.bz2
?

It's not the 492Mb of each file I mind, it's the idea of spending 12-18 hours building one to find I wanted the other one!

The only manual I can find for Festival is here:
http://www.cstr.ed.ac.uk/projects/festival/manual/
Dated 1999, for version 1.4

I sometimes feel like I've missed the basics somewhere.
Were it not for threads like this I'd be completely lost!

[mrplow]

you think 12-18 hours is bad, I'm currently compiling HTS-demo_CMU-ARCTIC-ADAPT and it's INSTALL says it should take 2-3 days! I'll let you know how it goes, I'm on 10.10 btw