- Synthesize Vietnamese speech with natural voices whose prosodics (duration, pause and amplitude) determined by a technique based on artificial neural networks.
- Support SAPI5 interface.
- Can work with text contain both Vietnamese and another language in Unicode format.
- Provide 2 voices "NHMTTS Voice (Male)" and "NHMTTS Voice (Female)".
- Allow binding another voice of non-Vietnamese language to a NHMTTS voice to support alouding text in both languages.
- Allow changing default break rate, duration rate, volume and pitch of NHMTTS Voices.
- Allow manipulating speech rate and volume on the fly.
- Highlight text and mimic mouth shape corresponding to the sound being synthesized.
- Demonstrates the engine with a simple TTSApp application.