is a site dedicated to news from fields including A.I., computational linguistics, robotics, developmental psychology, machine learning, and cognitive science, with special focus on language-related technologies.

This site also provides information about Experience-Based Language Acquisition (EBLA), the software system that I developed as part of my dissertation research at the LSU Department of Computer Science.

Brian E. Pangburn
May 27, 2003

Informal Memorial for Dr. Steve Seiden

There has been an informal memorial planned for Dr. Steve Seiden. The gathering will be in room 145 of Coates Hall (please note that the room was changed from 169 in order to accommodate more people) on Monday, June 17th at noon (12pm).

At the memorial everyone will have an opportunity to share their memories of Dr. Seiden. There will also have a moment of silence to reflect on this tragedy that has occurred.

To make a donation for Dr. Seiden's wife and newborn son, click here.

Birth Announcement: Steven Benjamin Richard Seiden

I learned this morning that Steven Benjamin Richard Seiden (son of the late Dr. Steve Seiden) was born last night (June 12th). He weighed in at 5 pounds 0.2 ounces and is 18 inches long. Both mother and baby are reportedly doing well.

Join the team for the project!

I have just created a team for the project. The team number is 935184859. If you are already a participant in the project, just click here to join the team (you'll need your account password). If you'd like to join the project, you can download the client program here.

If this is all new to you, the client allows you to use the many unused CPU cycles on your computer to help solve computational problems that are too large to be solved by a single computer in a reasonable amount of time. To find out more, click here.

Oh, and did I mention that you can win money?

Natural Language Software Registry

This morning I discoverd the Natural Language Software Registry (NLSR).

From their site...

The Natural Language Software Registry (NLSR) is a concise summary of the capabilities and sources of a large amount of natural language processing (NLP) software available to the NLP community. It comprises academic, commercial and proprietary software with specifications and terms on which it can be acquired clearly indicated.


I probably won't be posting much for the next two weeks because my wife and I have to take our son Jack to Dallas to have craniofacial surgery for his metopic craniosynostosis. Please keep him in your thoughts on Tuesday, May 21st.

You can see a few updated photos of Jack & Jeremiah here.

Update (5-21-2002): Jack had his surgery this morning and everything went extremely well. Thanks to all who were thinking about him. We were able to snap off a few photos before the swelling got too bad. You can view one of his face here and one of his scar here. It is amazing how much he looks like his brother now! It will be a rough few days (weeks) for us, but so far so good.

Update (6-7-2002): Jack has made an amazing recover and looks great. The boys are seven months old today. I've uploaded some new photos of them here and here.

It works! It really works!

Just wanted to post a quick update on the E.B.L.A. system. Yesterday I finished the bulk of the coding and began testing. On my first test set of seven animations, it successfully detected all of the entities (objects & relations) and resolved all of the lexemes describing each animation to the corresponding entities. It worked correctly in the presence of noise (i.e. adding words like "the" to the event descriptions). It was also able to resolve multiple word senses (i.e. there were two types of arms used in the animations).

More On IBM...

While I'm on the topic of IBM and language, I thought I should add a link to their ViaVoice speech recognition software page. This is the link for the ViaVoice developers site.

IBM's Work With Speech Recognition

This article details the work that IBM Research is doing to improve speech recognition.

This is another good article detailing IBM's work with speech recognition, and this is the link for IBM's Human Language Technologies Group.

Mechanical Rather Than Digital Voice Synthesis

Again from, comes a link to this story from New Scientist on Hideyuki Sawada's attempt to build an entire artificial voice system. Rather than using traditional digital voice synthesis to help robots speak, he is building a mechanical replica of the human voice system.

The article includes a diagram of the system and links to two MPEG videos of the system in action.

While not complete, it looks like Sawada has made a lot of progress.

From, Denis Susac, the former maintainer of, has launched his own AI site at There are tons of excellent resources on the site. Definitely worth a visit!

Syndicate content