[hts-users:00412] Re: adaptation using HTS

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

Hi,

Sarah Creer wrote:

> For this reason it seems that an adaptation method, that I have read about in some of the HTS group's publications, 
> would be more suitable. 
> From previous postings I read that HTS does not yet support speaker adaptation using HEAdapt so 
> is this still currently true? 

Yes and no.

Last week some members of HTS working group discussed about the final HTS-2.0 release
and we decided to include the speaker adaptation in this version.
Actually, some modifications to support the speaker adaptation for MSD-HMMs have already been made in HTS-2.0RC2.
Current implementation *probably* works, but it should be tested further and some essential functions have not been included yet.
I'll work on it this October, I hope I can it in HTS-2.0RC3.

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://kt-lab.ics.nitech.ac.jp/~zen
------------------------------------------------