Hello everybody,
I appreciate the help you've given me.
Sebastien, I make the Esther's question my question too.
Which was the rename table that you've used ?
How did you adapt the HTS-demo files to Festvox file format ? It was
necessary any conversion little/big endian ?
HTS-demo --> Festvox files
----------------------------------------------
dur.pdf --> duration.pdf
lf0.pdf --> lf0.pdf
???? --> mcep.pdf
???? --> p_mcep.pdf
label.feats --> feat.list
trees-dur.inf --> trees-dur.inf
trees-lf0.inf --> trees-lf0.inf
???? --> trees-mcep.inf
???? --> lf0_acc.win
???? --> lf0_dyn.win
???? --> mcep_acc.win
???? --> mcep_dyn.win
Thank you.
Gustavo Teodoro Laureano
On Wed, Feb 11, 2009 at 2:39 PM, Sebastien LE MAGUER
<Sebastien.Le_maguer@xxxxxxxx <mailto:Sebastien.Le_maguer@xxxxxxxx>>
wrote:
Hello,
in fact, I've juste updated the "nitech_us_slt_arctic_hts.scm"
file in the "nitech_us_slt_arctic_hts::hts_engine_params"
section. I've add the missing lines and rename the files.
Sebastien
Esther Klabbers a écrit :
Do you have any information on how you adapted those festvox
files? I've tried doing that in the past but have never gotten
it to work. There is also no documentation available about this.
Esther
On Feb 11, 2009, at 7:51 AM, Sebastien LE MAGUER wrote:
Hello,
when I used the demo, I modify the festvox files present
in the nitech_us_slt_arctic_hts voice to adapt them to the
demo. The directory hts is the directory contains the
files present in /voices/qst001/ver1.
Maybe I'm wrong but it worked for me.
Sébastien Le Maguer
Gustavo Laureano a écrit :
Hello everybody,
Is there a way to train a hts voice for festival
directly, or convert hts voice to festival voice ?
I have downloaded the
festvox_nitech_us_slt_arctic_hts-2.1.tar.bz2, but
theirs files apparently are not the same trained by
HTS-demo-SLT.
Is there a way to convert hts voice to festival voice?
In the message:
http://hts.sp.nitech.ac.jp/hts-users/spool/2007/msg00036.html
there is a file: hts-festival-2.0.tar.gz
<http://nshmyrev.narod.ru/temp/hts-festival-2.0.tar.gz>
to be downloaded. With these files can I use the voice
files created by hts-demo-slt ?
Thanks.
--
Gustavo Teodoro Laureano
Esther Klabbers-Judd
Assistant Professor,
Center for Spoken Language Understanding (CSLU),
Division of Biomedical Computer Science (BMCS)
Oregon Health & Science University (OHSU)
20000 NW Walker Road / Beaverton, OR 97006
Office: +1-503-748-3005 / Fax: +1-503-748-1306
http://www.cslu.ogi.edu/~klabbers
<http://www.cslu.ogi.edu/%7Eklabbers>
--
Gustavo Teodoro Laureano