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

[hts-users:00355] Re: about " decision trees"


Hi,

lei liu wrote:

I  found  there is only one tree for  duratoin ( state 2) ,
but for f0 and mcep , there are 5 trees (state 2 ,3 4, 5, 6 ) .

why?

Meaning of "states" is completely different between in HMMs and duration models.
Please read
Takayoshi Yoshimura, Keiichi Tokuda, Takashi Masuko, Takao Kobayashi and Tadashi Kitamura, ``Simultaneous modeling of spectrum, pitch and duration in HMM-based speech synthesis,'' Proceedings of European Conference on Speech Communication and Technology, Budapest, Hungary, vol.5, pp.2347-2350, Sep. 1999.

This paper is available at the publication page of HTS website.

why  the  index are all the negative?

This is based on HTK format.
So please ask this question to developers of HTK.

There is only one tree for every state in the "inf" file.
does it mean the tree is used by all the phoneme.

Yes.

Before i read the "inf"  file , I  thought that   there  is a tree for  one  phoneme.

No.
For speech recognition we should discriminate phonemes because mixing phonemes will degrade speech recognition performance.
However, for speech synthesis we can less care about it.

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


References
[hts-users:00354] about " decision trees", lei liu