Major flite update. Changelog:

New in 2.2 (Oct 2018)

• Better grapheme support (Wilderness Languages) hundreds of new languages

New in 2.1 (Oct 2017)

• Improved Indic front end support (thanks to Suresh Bazaj @ Hear2Read)
• 18 English Voices (various accents)
• 12 Indian Voices (Bengali, Gujarati, Hindi, Kannada, Marathi, Panjabi Tamil and Telugu) usually with bilingual (with English) support
• Can do byteswap architectures [again] (ar9331 yun arduino, zsun etc)
• flitecheck front-end test suite
• grapheme based festvox builds give working flitevox voices
• SAPI support for CG voices (thanks to Alok Parlikar @ Cobalt Speech and Language INC)
• gcc 6.x support
• .flitevox files (and models) 40% of previous size, but same quality

New in 2.0.0 (Dec 2014)

• Indic language support (Hindi, Tamil and Telugu)
• SSML support
• CG voices as files accessilble by file:/// and http:// (and set of 13 voices to load)
• random forest (multimodel support) improves voice quality
• Supports diffrent sample rates/mgc order to tune for speed
• Kal diphone 500K smaller
• Fixed lots of API issues
• thread safe (again) [after initialization]
• Generalized tokenstreams (used in Bard Storyteller)
• simple-Pulseaudio support
• Companion multilingual ebook reader Bard Storyteller https://github.com/festvox/bard

New in 1.4.1 (March 2010)

• better ssml support (actually does something)
• better clunit support (smaller)

New in 1.4 (December 2009)

• crude multi-voice selection support (may change)
• 4 basic voices are included 3 clustergen (awb, rms and slt) plus the kal diphone database
• CMULEX now uses maximum onset for syllabification
• alsa support
• Clustergen support (including mlpg with mixed excitation) But is still slow on limited processors
• Speed-ups in feature interpretation limiting calls to alloc
• Speed-ups (and fixes) for converting clunits festvox voices