FreeTTS

This release adds improvements to the emacspeak server demo, performance
improvements for time-to-first-sample and cancel, improvements to the
state name expansion logic, pronunciations for the British Sterling
and "not" signs, new pronunciations for Linux terms, a fix for a
-dump{Multi}Audio problem in the freetts application, elimination of a
hang with JDK1.5, tools specifically for importing CMU ARCTIC Voices, a
fix for MBROLA phoneme remapping, and rudimentary internationalization
for FestVox importation.

This release includes a number of minor
enhancements and bugfixes over the last release.
Enhancements include improved support for
importing FestVox voices, better support for
custom pronunciations, better support for
redirecting of audio, and improved compatibility
for the GNOME speech API.

This release is feature-equivalent to Flite 1.2, including the ability to import voice data directly from FestVox voices. GNU make has been abandoned in favor of Ant. This release uses jar file manifests to help reduce the need for a lengthy CLASSPATH, adds the ability to set the AudioPlayer on a JSAPI Voice, adds support for MBROLA on the Mac, and extends the voice packaging scheme to simplify management and addition of new voices, including dynamic detection and loading of voice jar files.

This version of FreeTTS includes support for the
MBROLA binaries, which gives FreeTTS a set of three
more voices. This release also includes a Java(TM)
WebStart example that demonstrates one way to use
FreeTTS and the Web to make applications speak.
Finally, a number of bugs have been fixed in this
release.

FreeTTS now includes a number of changes since 1.0. This release includes an improved front-end tokenizer, and a workaround for some sound problems on Linux. It is based upon Flite 1.1.