The outline will be

1.research trends (including Aurora and EARS projects) 
industry trends (telephony service, speech servers, 
voicexml vs.
2.salt, w3c grammar, speech sdk, xml, distributed SR, 
language coverage)
3.open sources
4.tools for source control

Bio.
     Dr. Hwang co-developed the SPHINX-II 
(http://fife.speech.cs.cmu.edu/sphinx/) speech recognition 
system, including acoustic training, at Carnegie Mellon 
University from 1987-1993 and pioneered the Markov state 
clustering work (senones) for hidden Markov modeling (HMM).

     After graduation, she joined the Speech Technology 
Group at Microsoft Research in April 1994, co-developing 
the Whisper dictation-oriented speech recognition system. 
Her research interests include pattern recognition 
(especially speech and handwriting recognition), statistic 
modeling, heuristic search, machine translation, natural 
language understanding, discrete math, compiler, and 
algorithms. In addition she is interested in database 
management and web design.

     With the system-oriented emphasis and the passion of 
putting the state-of-the-art technologies into end users' 
hands, she shifted from research to the product development 
in 1999. Involved in Microsoft Speech API and Office XP, 
she was responsible for the acoustic models for speech 
recognition and various related tools, including the HTK 
Toolkit. She is currently working on improving the 
recognition accuracy of Speech.Net through various desktop 
and telephony projects.