Subject: HMM-Based Speech Synthesis Toolkit version 1.1
From: Keiichi Tokuda 
To: hts: ;
Cc: tokuda@ics.nitech.ac.jp
Date: Mon, 12 May 2003 20:28:32 +0900 (JST)
X-Mailer: Mew version 2.3 on Emacs 20.7 / Mule 4.1 ()

This BCC email is to let you notice the new version release of
our software:

          The HMM-based Speech Synthesis System (HTS)
               version 1.1 release May 9, 2003

for HMM-based speech synthesis.  We apologize if you receive
multiple copies.  Please check

   http://hts.ics.nitech.ac.jp/

A brief expiation is attached bellow.

We would appreciate it if you would distribute this email to
anyone who might be interested in this software.

Best regards,
Keiichi Tokuda
tokuda@ics.nitech.ac.jp
http://kt-lab.ics.nitech.ac.jp/~tokuda/


****************************************************************

            HMM-Based Speech Synthesis System (HTS)
                Version 1.1 release May 9, 2003

The HMM-Based Speech Synthesis System (HTS) has being developed
by the HTS working group and others (see "Who we are" and
"Acknowledgments" in "README" file).  The basic core system of
HTS was implemented as a modified version of HTK
(http://htk.eng.cam.ac.uk/) together with SPTK
(http://kt-lab.ics.nitech.ac.jp/~tokuda/SPTK/), and is released
as HMM-Based Speech Synthesis System (HTS) version 1.1
(http://hts.ics.nitech.ac.jp/) in a form of patch code to HTK
version 3.2.  The patch code is released under a free license,
without commercial restrictions.  However, it should be noted
that once you apply the patch to the HTK source code, you must
obey the license of HTK.

The current version does not include any text analyzer but the
Festival Speech Synthesis System
(http://www.festvox.org/festival/) could be used as text
analyzer.  This distribution includes a demo script using "CSTR
US KED Timit" ( http://www.festvox.org/dbs/dbs_kdt.html) and
"CMU Communicator KAL limited domain"
(http://www.festvox.org/dbs/dbs_com.html).  Note that the
synthesized speech quality is not so good for "CMU Communicator
KAL limited domain" because the speech database was designed for
limited domain (not for general purpose).  Other synthesized
speech samples can be found at
http://kt-lab.ics.nitech.ac.jp/~tokuda/HTS-demo/.  An
explanation of techniques and algorithms used in the HMM-based
speech synthesis system can be found at
http://kt-lab.ics.nitech.ac.jp/~tokuda/HTS-demo/synHMM_o_e.pdf.

HTS version 1.1 comes with a small run-time synthesis engine
(less than 1 M byte including acoustic models), which can run
without the HTK library.

Two HTS voices for Festival trained by using 450 utterances of
"CSTR US KED Timit" ( http://www.festvox.org/dbs/dbs_kdt.html)
and 523 utterances of "CMU Communicator KAL limited domain"
(http://www.festvox.org/dbs/dbs_com.html), respectively, are
also released with HTS version 1.1.  They are based on the small
synthesis engine.  Each of HTS voices can be used as a "voice"
of Festival Speech Synthesis System without any other HTS tools.

****************************************************************

What's new in version 1.1

 - A small synthesis engine (to be called from Festival). 
 - HMM file format converter for the engine. 
 - Many fixed bugs (Thanks for reporting them). 
 - Accompanied by HTS voices for Festival. 

****************************************************************