This page represents the confluence of my work on A.I. and Irish Heritage. It details all the software systems I'm developing for Irish Heritage.
Using Koha to computerize this library's operations.
Here is a link to a video of a robotic book scanner I built practicing turning the pages of the text: "Learning Irish (Mícheál Ó Siadhail)". If I hadn't destroyed the robotic scanner during a psychosis (thinking that the spread of information was the Devil), I had planned to scan many of the books at the IAHC Library (http://www.irishamhc.com).
The goal is to combine scanned pages with a free OCR system such as Tesseract or this other one Google is developing (I forget the name). This project is obviously more general than Irish history but Irish history is probably my greatest or second greatest motivation for doing so. Check out the Universal Library. Maybe it would be better simply to loan them our collection?
This is to be a collection of libre Irish music, meaning there are no restrictions on its redistribution.
Knowledge Archives. Wrote some software that tags entities in text and makes Wiki pages out of them with links to the containing sentences.
Tara: I've set up several online services, such as a chat room, a letter template, an issue tracker, a myspace scraper and contact agent, and daily reminders by text message, all to support Tarawatch and similar campaigns
Here are things that are no longer necessary, because they have been completed by others.
I wrote a simplistic Text to Speech system for Irish. Note that it currently doesn't select the best pronunciation, only the first.
I am working with Apertium to develop English to Irish MT. However, as I'm unsure of the political consequences of such a tool, and per some advice I have received, I am holding off on this for now.
Creating a Gutenberg etext and high res scans of the text as there doesn't seem to be a freely available English translation of the text, even though it is out of copyright. More texts hopefully to follow.