Thanks you, Heiga.
I want to try GV function to predict prosodic parameters, and it seems that there are two methods to approach the goal:
1. to port GV function to hts engine;
2. to modify HMGenS toolkit to generate prosodic parameters as well as hts engine;
Are both methods reasonable? if yes can you give me some suggestion to achieve the target?
On Dec 26, 2007 9:40 PM, Heiga ZEN (Byung Ha CHUN) <
zen@xxxxxxxxxxxxxxx> wrote:
Hi,
Yongguo Kang wrote (2007/12/26 22:25):
> I have a question about GV. As we know, the new algorithm of GV
> has been added into HTS 2.1alpha, but I cannot find any code in the
> latest hts engine. Meanwhile, there are some codes containing GV in
> HMGenS module. So how can I use GV in the hts engine to predict
> prosodic parameters? or directly use HMGenS module ?
hts_engine has not supported GV yet.
We are planning to include it in the final stable release, hts_engine API ver.
1.0.
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
------------------------------------------------