[hts-users:00111] Re: How can I check if mcep is well extracted..

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

Hi,

sangjin kim wrote:

> 1. I'm gonna utilize the format of English. Can I add or disable some
> part of the format? For example, in the English format, can I disable
> /D: /E: /F: option? or can I add /K: /L: /M: option easily?

Yes.

> 2. Can I disable some part of one option? For example, in the English
> format, can I disable b1 b2 b3 in /B: option? ( just for example.. )

Yes.

> In case of YES, which code should I modify? 

First of all, define your context-dependent label format for your language.
Then, generate context-dependent label sequence for each training data.
Questions used in decision tree-based context clustering have to be 
modified to fit your own label format.
After training HMMs using them, convert input text into 
context-dependent label sequence using your converter.

Training part of HTS is completely independent of languages.
So you don't need to modify any HTS source code.

Please modify questions, your converter, and context-dependent labels of 
your corpus.

Regards,

Heiga ZEN (Byung Ha CHUN)

--
 ------------------------------------------------
  Heiga ZEN     (in Japanese pronunciation)
  Byung-Ha CHUN (in Korean pronunciation)

  Department of Computer Science and Engineering
  Graduate School of Engineering
  Nagoya Institute of Technology
  Japan

  e-mail: zen@xxxxxxxxxxxxxxxx
     web: http://kt-lab.ics.nitech.ac.jp/~zen
 ------------------------------------------------