Festival Source Distribution
version 1.95 (2.0 beta): July 2004Festival offers a general framework for building speech synthesis systems as well as including examples of various modules. As a whole it offers full text to speech through a number APIs: from shell level, though a Scheme command interpreter, as a C++ library, and an Emacs interface. Festival is multi-lingual (currently English, Welsh and Spanish) though English is the most advanced.
The system is written in C++ and uses the Edinburgh Speech Tools Library for low level architecture and has a Scheme (SIOD) based command interpreter for control. Documentation is given in the FSF texinfo format which can generate, a printed manual, info files and HTML.
This distribution includes:
- Full English (British and American) text to speech
- Full C++ source for modules, SIOD interpreter, and Scheme library
- Lexicons based on CMULEX and OALD (distributed with permission)
- Edinburgh Speech Tools, low level C++ library
- HTS hidden markov model based synthesis engine
- 4 HTS American English Voices (From CMU)
- Multisyn general purpose unit selection engine
- Various multisyn voices (More to follow)
- British English diphone database (for residual LPC resynthesis) (8k (2.1M) and 16k (7.5M) versions of the database are included)
- 2 American English diphone databases (for residual LPC resynthesis) (8k (3.0M) and 16k (5.6M) versions of the database are included)
- 4 voices (1 British male, 2 American male and 1 American female) using the MBROLA system. (The mbrola program and voice databases are not distributed fromt his site).
- Castilian Spanish diphone database (for residual LPC resynthesis)
- Full documentation (html, postscript and GNU info format)
Festival version 1.95 is available for free for unrestricted use, see here for copyright.
- Download from CSTR (Primary UK site)
- Download from festvox.org (Primary US site)
- Information regarding the installation of MBROLA voices is here.
- Preliminary versions of the Multisyn build tools are available here
4 American voices and Mexican Spanish are available with a signal processing module from the TTS groups at CSLU at Oregon Graduate Institute. http://cslu.cse.ogi.edu/tts/download
Older versions of Festival have been removed from distribution to save space. However, if you need an older version for some reason please mail festival-help@cstr.ed.ac.uk and we will make it available for you
Requirements
To run Festival you need:
- A Unix machine: Festival has compiled and run on Suns (SunOS and Solaris), FreeBSD, Linux, SGIs, HPs and DEC Alphas but should be portable to any standard Unix machine. Festival will also compile on MS Windows platforms given a little work and patience
- A C++ compiler, we have tested the following:
- Sun Sparc Solaris 2.8/2.9: gcc-2.95.3, gcc-3.2 gcc-3.3
- Intel Solaris 2.5.1: GCC 2.7.2
- FreeBSD for Intel 3.x, 4.x: GCC 2.95.x, GCC 3.0
- Linux 2.x for Intel (including RedHat 6.[012],.7[01]): GCC 2.7.2, GCC 2.7.2/egcs-1.0.2, egcs-1.1.2, GCC 2.95.[123], GCC "2.96", GCC 3.0
- Windows NT 4.0, Windows95, Windows 98: GCC with egcs (Cygwin 1.1), Visual C++ 6.0.
- GNU Make any recent version
- Audio hardware, /dev/audio (8bit and 16bit for Suns, Linux and FreeBSD) and NCD's NAS network transparent audio system are supported directly but Festival supports the execution of any Unix command that can play audio files.
Restrictions
Unlike versions prior to 1.4.0 this version is free for any purpose, commercial or non-commercial alike. Although it is our intention that the copyright allows commercial use without further licensing if you intend to use this commercially please check the actual copyright and the COPYING file in each of the distributions. If you do use the system commercially we would be interested to know so we can show we are making a useful contribution.
Note that the British English lexicon is derived from Oxford Advanced Learners' Dictionary of Current English and that sub-system alone in our distribution is restricted for non-commercial use only. We hope to rectify this with an alternative free lexicon in the near future.
Thank you for your continued interest in our work.