[hts-users:00417] Re: Unseen triphone models
- Subject: [hts-users:00417] Re: Unseen triphone models
- From: "Heiga ZEN (Byung Ha CHUN)" <zen@xxxxxxxxxxxxxxxx>
- Date: Sat, 16 Sep 2006 05:30:09 +0900
Hi,
Zuko Fani wrote:
I checked the training script and found the full_all.list had models
that were not in my mono.list. I saw that the models were coming from
the Alice0*.lab files.
Because alice0*.lab files were designed for ARCTIC database.
If you are planning to use hts_engine rather than HMGenS to generate speech, you need not to make unseen models.
Please set
$MKUN1 = $PGEN1 = $WGEN1 = $UPMIX = $ERST6 = $CXCL5 = $MKUN2 = $PGEN2 = $WGEN2 = 0
in Config.pm and use hts_engine with files in voices directory.
But am struggling to read how to construct my own
coz there are some character formats that are not documented on your
lab_format.pdf for example the sequence of numbers and characters after
the triphones.
l@x_x/A:0_0_0/B:x-x-x@x-x&x-x#x-x$x-x!x-x;x-x|x/C:1+1+2/D:0_0/E:x+x@x+x&x+x#x+x/F:content_2/G:0_0/H:x=x@1=10|0/I:19=12/J:79+57-10
and
d@1_3/A:0_0_2/B:1-0-3@1-1&3-8#1-7$1-2!2-1;2-7|ae/C:1+0+4/D:content_2/E:aux+1@2+7&2+4#1+1/F:content_1/G:3_3/H:10=8@2=8|L-L%/I:8=6/J:63+49-9
Sorry I cannot understand what your question means.
Regards,
Heiga Zen (Byung Ha Chun)
--
------------------------------------------------
Heiga ZEN (in Japanese pronunciation)
Byung Ha CHUN (in Korean pronunciation)
Department of Computer Science and Engineering
Nagoya Institute of Technology
Gokiso-cho, Showa-ku, Nagoya 466-8555 Japan
http://kt-lab.ics.nitech.ac.jp/~zen
------------------------------------------------
- Follow-Ups
-
- [hts-users:00418] Re: Unseen triphone models, Luis Uebel
- References
-
- [hts-users:00416] Re: Unseen triphone models, Zuko Fani