[hts-users:00392] Re: Questions about multi-mixture parameter generation

["Heiga ZEN (Byung Ha CHUN)" <zen@xxxxxxxxxxxxxxxx>]

Hi

sxu wrote:

> 1. With a given state sequence q and state durations, how to get the mixture sequence i that maximizes P(O|Q,lambda)?
> 3. Is there a paper describe the algorithms about multi-mixture parameter generation?

Please read 

K. Tokuda, T. Masuko, T. Yamada, T. Kobayashi, S. Imai,
"An algorithm for speech parameter generation from continuous mixture HMMs with dynamic features,"
in Proc. Eurospeech, pp.757-760, 1995.

> 2 Is there a version of HTS that supports the generation from 
> multi-mixture?

HTS-2.0RC2 does.

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