[hts-users:00464] Re: Speech parameter Generation
- Subject: [hts-users:00464] Re: Speech parameter Generation
- From: "Heiga ZEN (Byung Ha CHUN)" <zen@xxxxxxxxxxxxxxxx>
- Date: Wed, 06 Dec 2006 18:41:39 +0900
Hi,
Zuko Fani wrote:
In your thesis Chapter 4 section 4.2.2 you have the mean vector as a
(3K x 1) and the covariance matrix (3K x 3K ).
But I see in HTS from the HMM files that the vector size of the mean is
(75 X 1). Does that mean that ( K = 28 ).
K = 25 (mcep)
K = 1 (lof F0)
And that the first three values belong to ( C, deltaC and deltadeltaC).
1-25: static mcep
26-50: delta mcep
51-75: delta^2 mcep
76: static log F0
77: delta log F0
78: delta-2 log F0
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
------------------------------------------------
- Follow-Ups
-
- [hts-users:00465] HTK question, Patrick Davin
- References
-
- [hts-users:00463] Speech parameter Generation, Zuko Fani