[hts-users:00547] Re: help in SPTK(Now HMGenS)

[Mohammed Samir <msamirhb@xxxxxxxxx>]

--- "Heiga ZEN (Byung Ha CHUN)" <zen@xxxxxxxxxxxxxxxx>
wrote:

> Hi,
> 
> Mohammed Samir wrote:
> 
> >> BTW, why do you want to use mlpg in SPTK?
> >> I think HMGenS in HTS-2.0 provides all functions
> >> which you need.
> >>
> > Currently I don't want to build a big project just
> I
> > want to try synthysise 3 or 4 phonemes only
> because I
> > don't have suffcient data.
> > my plane was:
> > 1-training using HTK3.3 by (HCopY, HInit and
> HRest).
> > 2- produce pdf sequence from the traind HMM using
> > HMGenS.
> > 3-generate parameters from this pdf sequence by
> > mlpg(of SPTK) .
> > 4- use mlsadf (SPTK) to synthesise using generated
> > parameters.
> > is that possible?
> 
> It is possible but even a small project, I do
> recommend you to use HMGenS 
> because it supports small model set and much more
> flexible than mlpg in SPTK.
> Furthermore, mlpg in SPTK is not maintained for a
> long time.
> 
> > this training (step 1) didn't include state
> durations,
> > How can I Add state durations in training to be
> fit
> > for HMGenS?
> 
> So which version of HTS are you using?
> If you are using HTS-2.0, HRest can generate state
> duration pdf by specifying -g option.
> If you don't, use HTS-2.0.
> 
> Regards,
> 
> Heiga ZEN (Byung Ha CHUN)
> 
> -- 
I follow your advice and HTS-2. has been installed
(Thank you). Now I have 5 utterance of 1 phoneme, I
have use HCopy to generate *.mfc files, then these
files introduced with proto file to HInit to initiate
HMM. Then I Used HRest with -g optin. Now I want to
produce sound file(wave or any signal) from this
output(  state duration pdf) can it be done using
HMGenS only? and How?


 
____________________________________________________________________________________
Don't pick lemons.
See all the new 2007 cars at Yahoo! Autos.
http://autos.yahoo.com/new_cars.html