• Jan 24th 2007: Blizzard Challenge 2007 The call for participation in the third Blizzard Challenge is posted. Blizzard is a multi-site challenge to build a synthetic voice from a common database that will be evaluated by a large number of listeners.
• Jan 21st 2007: Festvox-2.1 Release Festvox-2.1 has been released. New with this release is Clustergen Statistical Parametric Synthesis support, EHMM acoustic labeler (no longer dependent on Sphinx) Thanks to Kishore, Voice Conversion support (thanks to Tomoki Toda), and cygwin support. See ANNOUNCE-2.1 for more details.
• Dec 29 2006: HMM-Based Speech Synthesis System HTS-2.0 has been released. HTS version 2.0 now supports adaptation and adaptive training based on MLLR. MAP-based adaptation is also supported. This version does not include any text analyzer yet, but you can use the Festival Speech Synthesis System as a text analyzer. This distribution includes demo scripts using CMU ARCTIC database (English). Six HTS voices for Festival 1.95 are also released. They are based on our small synthesis engine which has been included as a module of Festival. Each of the HTS voices can be used without any other HTS tools.
Sep 06: The Blizzard Challenge 2006 consisted on 14 different systems, results of the evaluation and the papers describing the systems are available here.
• June 04: 5th ISCA Speech Synthesis Workshop June 14-16th 2004, Carnegie Mellon University: Proceedings

The Festvox project aims to make the building of new synthetic voices more systemic and better documented, making it possible for anyone to build a new voice. Specifically we offer:

• Documentation, including scripts explaining the background and specifics for building new voices for speech synthesis in new and supported languages.
• Specific scripts to build new voices in supported languages, such as US and UK English.
• Aids to building synthetic voices for limited domains
• Example speech databases to help building new voices.
• Links, demos and a repository for new voices

• Better clunits general voice support
• Support for CMU Sphinx and SphinxTrain to build acoustic models for labeling
• DOCBOOK version of the documentation, with more general backgfround documentation
• Initial support for Mac OS X
• configure support to match Edinburgh Speech Tools

This work is firmly grounded within Edinburgh University's Festival Speech Synthesis System and Carnegie Mellon University's small footprint Flite synthesis engine

This work has been supported be various groups including, Carnegie Mellon University, the US National Science Foundation (NSF), and US Defense Advanced Research Projects Agency (DARPA).

You will need:

• To read the documentation
• A Unix machine (e.g. Linux, FreeBSD, Solaris, etc) with working audio i/o. This may work on other platforms but many scripts, perhaps unnecessarily, depend on Unix utilties like, awk, sed etc.
• Installed versions of Edinburgh University's Festival Speech Synthesis System and Edinburgh Speech Tools (distributed with Festival).
• A waveform viewing/labeling program like emulabel distributed as part of Macquarie University's EMU speech database system. Although automatic labeling software is included in festvox, a display tool is necessary for diagnosis and debugging.
• Patience and care, and a little interest in the subject of speech technology.