[Subject Prev][Subject Next][Thread Prev][Thread Next][Date Index][Thread Index]

[hts-users:01061] Re: Problem about HMGenS


Thanks a lot ! I'm so sorry to post the same message three times  and I have another question:

    If I want to reduce the vector size of delta, delta-delta,e.g. original feature size is 25,but delta is 10,and delta-delta is 15,
    1)
how could I design  the window ,Is there any method?
    2) do I just modify Calc_WUM_and_WUW() to support the window to achieve my goal?
Thank you,It's very kind of you , Heiga.
 2007/12/18, Heiga ZEN (Byung Ha CHUN) <zen@xxxxxxxxxxxxxxx>:
Hi,

# Do not post the same message three times, please.
# And I think the subject of this post is wrong, because it is not a problem of HMGenS.

wei chen wrote (2007/12/17 22:26):

> 1) the vector size of delta, delta-delta and original features
> must be same, e.g., the size of spectrum is 25 in hts demo?

In HTS-demo, yes.

> 2) Is it possible to reduce the vector size of delta, delta-delta, e.g. original feature size is 25,
> but delta is 10,and delta-delta is 15?

Yes.

> 3) if the 2nd assumption is allowed, how can we modify HMGenS
> algorithm?

Please modify HGen.c:Calc_WUM_and_WUW() to support your window.

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



Follow-Ups
[hts-users:01062] Re: Problem about HMGenS, Heiga ZEN (Byung Ha CHUN)
References
[hts-users:01057] Problem about HMGenS, wei chen
[hts-users:01060] Re: Problem about HMGenS, Heiga ZEN (Byung Ha CHUN)