[hts-users:01489] Re: hts_engine-0.99 does not generate normal speech in HTS-demo using 12-th order LSP
- Subject: [hts-users:01489] Re: hts_engine-0.99 does not generate normal speech in HTS-demo using 12-th order LSP
- From: "Heiga ZEN (Byung Ha CHUN)" <zen@xxxxxxxxxxxxxxx>
- Date: Fri, 27 Jun 2008 01:14:40 +0900
- Delivered-to: hts-users@xxxxxxxxxxxxxxx
Hi,
Sacha Krstulovic wrote (2008/06/27 1:17):
In the failing HTS_calloc call the values of num and size are actually
num=1060370154 (GASP!, that's why calloc fails) and size=4.
I see that you have changed the GV pdf loading mechanism
between hts_engine_API-0.95 and -0.96/0.99, and besides I was
trying to use models trained with HTS-2.1beta into hts_engine-0.96/0.99.
Did you change the GV pdf format between HTS-2.1beta and HTS-2.1RC1?
Yes, we changed the header of GV pdf in hts_engine API 0.96.
In 0.95, its format is
dimensionality (4-byte integer)
mean (float)
...
variance (float)
...
In 0.96, its format is
dimensionality (4-byte integer)
number of pdfs (4-byte integer)
mean (float)
...
variance (float)
...
Similar discussions have been made in hts-users:01300, 1302, 1303, and 1304.
If yes, does that mean that although the hts_engine_API and HTS
are distributed as separate packages and with different version
numbering schemes, there is still a strong mutual dependency
between precise versions of both packages?
GV pdfs are generated by HTS-demo/data/Makefile, not HTS itself.
So I think hts_engine API and HTS are independent.
Sometimes we have to change model format etc.
I recommend you to re-train your model using the latest HTS and demo script
because the latest one may have less bugs.
# I personally believe that the latest HTS is always the best one :-)
Best 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://www.sp.nitech.ac.jp/~zen
------------------------------------------------
- Follow-Ups
-
- [hts-users:01490] Re: hts_engine-0.99 does not generate normal speech in HTS-demo using 12-th order LSP, Sacha Krstulovic
- References
-
- [hts-users:01462] hts_engine-0.99 does not generate normal speech in HTS-demo using 12-th order LSP, Xuchen Yao
- [hts-users:01463] Re: hts_engine-0.99 does not generate normal speech in HTS-demo using 12-th order LSP, Heiga ZEN (Byung Ha CHUN)
- [hts-users:01483] Re: hts_engine-0.99 does not generate normal speech in HTS-demo using 12-th order LSP, Heiga ZEN (Byung Ha CHUN)
- [hts-users:01484] Re: hts_engine-0.99 does not generate normal speech in HTS-demo using 12-th order LSP, Sacha Krstulovic
- [hts-users:01485] Re: hts_engine-0.99 does not generate normal speech in HTS-demo using 12-th order LSP, Heiga ZEN (Byung Ha CHUN)
- [hts-users:01488] Re: hts_engine-0.99 does not generate normal speech in HTS-demo using 12-th order LSP, Sacha Krstulovic