From: Heiga Zen/Byung-Ha Chun <zen@xxxxxxxxxxxxxxxx>
Reply-To: hts-users@xxxxxxxxxxxxxxxxxxxxxxxxx
To: hts-users@xxxxxxxxxxxxxxxxxxxxxxxxx
Subject: [hts-users:00056] Re: help
Date: Sat, 17 Jul 2004 13:12:16 -0400
Hi,
yi cindy wrote:
Sorry for replying too late.
> The "readme" downloaded from
> http://hts.ics.nitech.ac.jp/download.html
> instructs all programs can be compiled and can work on LINUX and
> UNIX workstations. However, how to comply and make them work
properly
> under Windows?
Please use Cygwin.
> The source code of HTS is downloaded from
> http://hts.ics.nitech.ac.jp/download.html.
> We can see the segment about speech synthesis.
> We can not find the segment about speech analysis.
Sorry, I cannot understand the meaning of this question.
In HTS, we train HMMs and generate speech from them.
> The demo from
> http://kt-lab.ics.nitech.ac.jp/~tokuda/HTS_demo/hmm_vocoder/
> show us the comparison of HTS of different bits/s.
> How to change the number of bits/s ?
Actually, speech samples of HMM vocoder on that site are based on
very
old system.
You can find the latest system on
"An HMM-based approach to speaker dependent 100bit/s speech coding,"
Special Workshop In Maui, 2004.
Please read it.
> If we want to run the program successfully, we should get and set
the
> environment config files
> (such as decision trees file for state duration
> Log F0,Mel-Cepstrum ;model file for state duration, Log
> F0,Mel-Cepstrum;window file for culcuration delta of log F0).
> However, how to get and set them?
Please run HTS-demo.
They will be generated at HTS-demo/voices.
Best regards,
Heiga Zen / Byung-Ha Chun
--
------------------------------------------------
Heiga Zen (in Japanese pronunciation)
Byung-Ha Chun (in Korean pronunciation)
Department of Computer Science and Engineering
Graduate School of Engineering
Nagoya Institute of Technology
Japan
e-mail: zen@xxxxxxxxxxxxxxxx
web: http://kt-lab.ics.nitech.ac.jp/~zen
------------------------------------------------