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

[hts-users:00550] Re: help in SPTK(Now HMGenS)


Hi

Mohammed Samir wrote:

 I thought that I can generate parameters from state
duration pdf and then use mlsa filter to synthesize
.(thats all what I want for this single phoneme)can it
be done?

If you are using mel-cepstral coefficients, yes.
If you are using MFCC, no.

also in this experement I didn't  use
dynamic feature

In that case you will get sequences of mean vectors.

and if I want i think that I will add
this option  in the config file input of HCopy.

I don't recommend you to use HCopy.
Use SPTK's mcep and window.pl included in the HTS-demo.
Please refer demo for detail.

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:00551] Re: help in SPTK(Now HMGenS), Mohammed Samir
References
[hts-users:00549] Re: help in SPTK(Now HMGenS), Mohammed Samir