[hts-users:00986] [RE] Re: Binary file size of HTS

["Han, Seungho" <space0128@xxxxxxxxx>]

Thank you.
 
And then, if the excitation model is modified (that means more excitation parameters are used),
how do we get those excitation parameters from the 'pdf' files to synthesis the speech by hts_engine?
 
In my guess, the souce code of hts_engine should be modified. Is it right?
 
On the other hand, do we get the excitation parameters only including the added parameters with current hts_engine version (with no modification) when the other excitation parameters are added ?
 
Best Regards,
 
Seungho Han.

 

-------------------------------- [ 원 본 메 세 지 ] -------------------------------

> 보낸사람(From) : "Heiga ZEN (Byung Ha CHUN)" < zen@xxxxxxxxxxxxxxx > 
> 받는사람(To) : hts-users@xxxxxxxxxxxxxxx 
> 보낸날짜(Date) : 2007-11-21 01:00:28 
> 제 목(Subject) : [hts-users:00970] Re: Binary file size of HTS 
>
>
> Hi,
>
> Han, Seungho wrote (2007/11/20 16:00):
>
> > I want to check the binary file size of the synthesis system after HMM 
> > training like Table 7.1 in Reference A.
> > (Reference A: K. Tokuda, H. Zen, and A.W. Black, "An HMM-based approach 
> > to multilingual speech synthesis," in Text to Speech Synthesis: New 
> > Paradigms and Advances, Chapter 7, pp. 135-153, Pretice Hall, 2004.)
> > However, I don't know exactly what files should be anlayzed.
> > Could you help me know the binary file size ?
>
> Please check files under voices/... directory.
> *.pdf files correspond to "Gaussian parameters" and *.inf files correspond to "decision tree" in that table. 
>
> 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
> ------------------------------------------------