Festival 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, from Java, and an Emacs interface.
This project is multi-lingual (currently English (British and American), and Spanish) though English is the most advanced. Other groups release new languages for the system. And full tools and documentation for build new voices are available through Carnegie Mellon's FestVox project (http://festvox.org)
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.
Festival is free software. Festival and the speech tools are distributed under an X11-type licence allowing unrestricted commercial and non-commercial use alike.
Here are some key features of "Festival":
· Backwards compatibility with Festival 1.4.3
· English (British and American), Spanish and Welsh text to speech
Externally configurable language independent modules:
· letter-to-sound rules
· part of speech tagging
· intonation and duration
· diphone based: residual excited LPC (and PSOLA not for distribution)
· MBROLA database support.
· distributed under a free X11-type licence
· generalisation of stats modules, ngram, CART, wfst with viterbi so they can be shared more easily
· Initial JSAPI support
· XML load for Relations
· Portable (Unix) distribution
· On-line documentation
· SABLE markup, Emacs, client/server (including Java), scripting interfaces.
What's New in This Release:
· HTS hidden Markov model based synthesis engine from Nagoya Institute of Technology.
· Multisyn general purpose unit selection synthesis engine from CSTR
· support for gcc 3.2 and 3.3, and intel 8.0
· clunits unit selection improvements
· Apple OS X support
· bug fixes galore.