[hts-users:00663] question about training new voice

["Ching-Hsien Lee" <chinghsienlee@xxxxxxxxx>]

Hello ,

i want to train my own voice of cmu arctic database, so i recorded the wav files, and i searched articles of ML

i still had some questions.

According to "Training a new voice" article(http://hts.sp.nitech.ac.jp/hts-users/spool/2006/msg00213.html )

in that :

----------------------------------------------------------------------------------------------------------------------------------------------

...

(2)    Downloaded two male voices

 a.    http://www.speech.cs.cmu.edu/cmu_arctic/packed/cmu_us_bdl_arctic-0.95-release.tar.bz2

 a.    http://www.speech.cs.cmu.edu/cmu_arctic/packed/cmu_us_rms_arctic-0.95-release.zip

(3)     Converted audio files from WAV format to RAW.

(4)     Copied utt files to the training directory

 ../festival/utts/*.utt à <training>/data/utts/

...

----------------------------------------------------------------------------------------------------------------------------------------------

 

the first question is that accroding above list (4)  "Copied utt files to the training directory"

how to generate *.utt ? does use festvox, building clunits voice units database to generate my *.utt ?

and besides the *.utt, what do the content of /HTS-demo_CMU-ARCTIC-SLT/data/  i need to generate myself ?

 

Thanks a lot!