Some pictures taken while playing with Raymond Tan's camera. *8=)
Since there are so few pictures I didn't bother to thumbnail them. Hope it doesn't take too long to load.
This is one of my favourite pictures, taken during Commencement 2003 when I had the rare privilege of serving as Chief Graduand Marshall. The two delightful ladies are Xueling (left) and Renny (right).
The "hearing aid" in my ear is connected to what is purported to be a walkie-talkie. Its main role is to transmit loud static into my ear during the ceremony.
This shot was taken by Phillip Lam on his very nice camera. Phillip is an excellent photographer. If you have a wedding or any function, give him $10,000 and ask him to snap some shots for you *;=).
Don't ask.
This time, soon-to-be-graduands! Wai See on the left, and Xuan Linh on the right.
From left: Penguin, Bear, Cat and Zoo-Keeper,
otherwise known as Eugene, Colin, Damit and Raymond.
A shot of my notebook screen. The fragment of code shown here is part of a huge
(30,000+ line) program that models belief and doubt in spoken-language understanding.
This part of the program takes a spoken utterance (e.g. "If Pedro owns a donkey he will
be rich") that has already been rendered into a DRS (Discourse Representation Structure. Again,
don't ask), evaluates the utterance against knowledge already stored, and returns the system's
degree of belief and doubt in the utterance.
Other parts of the program would take in an utterance from a speech recognition engine (Sphinx-II from
CMU), clean up the utterances (removing umms, aaahs, false starts), use statistical measures and logical
coherence algorithms to correct mis-recognized words, render utterances into DRSs and BAFs (Belief
Augmented Frames), perform conflict detection and resolution against existing rules and knowledge, and update
existing rules and thematic structures.
As I've said; the program is huge.