Befejeztem az öngyógyító kódvázat (szakzsargonban: TDD TCR repó). Közzétettem az AI NLP kurzus előértékesítési oldalát. Készítettem napi hétköznapi reel-eket. Élveztem egy nagyszerű emelkedős sprintsorozatot az Olimpiai stadionnál.
Kategória: Uncategorized
-
More Indigenous language knowledge, less diabetes
Did you know? After adjustment for socio-economic factors, diabetes is significantly less prevalent in communities with more cultural continuity as measured by traditional Indigenous language knowledge (Oster et al., 2014, p. 9).
Work cited:
Oster et al.: Cultural continuity, traditional Indigenous language, and diabetes in Alberta First Nations: a mixed methods study. International Journal for Equity in Health 2014 13:92. doi:10.1186/s12939-014-0092-4
-
Presentation at HCI International 2022
Looking forward to speaking at the HCI International 2022 Conference in Gothenburg, Sweden today and presenting my new paper Say It Right: AI Neural Machine Translation Empowers New Speakers to Revitalize Lemko during session S143: Artificial and Augmented Intelligence Applications on Language Text and Speech Related Tasks.
Paper in conference proceedings: https://link.springer.com/chapter/10.1007/978-3-031-05643-7_37
Please cite as:
Orynycz, P. (2022). Say It Right: AI Neural Machine Translation Empowers New Speakers to Revitalize Lemko. In: Degen, H., Ntoa, S. (eds) Artificial Intelligence in HCI. HCII 2022. Lecture Notes in Computer Science(), vol 13336. Springer, Cham. https://doi.org/10.1007/978-3-031-05643-7_37 -
New Experiment: Lab-Made Lemko?
I will be conducting an experiment this month to see if machines can be made to translate into Lemko better than Google Translate or humans.
Hypothesis
A machine can be configured to translate from English into the endangered Slavic language of Lemko and achieve quality scores higher than those of Google Translate’s Ukrainian service, but not yet higher than those of humans.
Predictions
- My English to Lemko rule-based machine translation (RBMT) engine will achieve a bilingual evaluation understudy (BLEU) score of 15 against a clean bilingual corpus.
- The above engine will achieve a BLEU score that is a third higher (e.g. 20) when coupled with an improvised dictionary-based machine translation (DBMT) created from Lemko-Polish unit-test assertion pairs.
- Google Translate’s English to Ukrainian translation service will achieve a BLEU score of 10 against the above corpus.
- I, a human, will achieve a higher BLEU score than all the above machines against the above corpus.
The experiments will be conducted over the next week or so, for subsequent publication.
-
AI Empowers Speakers of Endangered Languages like Lemko Event Watchable On-Demand
Sharable Link:
https://www.orynycz.com/lemko/watch-ai-empower-new-speakersThe October 26, 2021 product launch and interactive demonstration was taped as can now be watched in its entirety here.
To watch and follow along, visit the event page.
Interactive Demo: AI Empowers Speakers of Endangered Languages like Lemko Sharable Link:
https://www.orynycz.com/lemko/watch-ai-empower-new-speakersYouTube and Facebook
You can also watch, like, comment, and share the presentation on Facebook or on YouTube.
-
Event: Artificial Intelligence Empowers Speakers of Endangered Languages like Lemko
October 26, 2021 | 6 p.m. EST
In a demonstration sponsored by the University of Pittsburgh
Center for Russian, East European, and Eurasian Studies, Pittsburgh
Nationality Rooms and Intercultural Exchange Programs, and University Center for International Studies, as well as the Carpatho-Rusyn Society, engineer and linguist Petro Orynycz unveils hybrid artificial intelligence technology that empowers new speakers of Lemko to read in the language immediately. Implications for endangered, low-resource language revitalization are discussed.See here for the official announcement on the website of the Nationality Rooms of the University of Pittsburgh.
Information can also be found on Facebook and LinkedIn.
Official Zoom link:
https://pitt.zoom.us/meeting/register/tJckdemprTMiGdClNhkIz9n1bexGm6ANZZkF
-
Lemko Artificial Intelligence Manhattan Project Breakthrough Delivers Humanlike Translations
MANHATTAN, Sept 25 (LEMKOTRAN.COM) – Breakthrough artificial intelligence (AI) neural machine translation (NMT) technology has delivered real-time translations at near-human quality, with Lemko to English engines deployed in Manhattan achieving a speed of 95,877 words per hour and corpus bilingual evaluation understudy (BLEU) score of 17.74 against a professional human linguist’s score of 28.
„These newly deployed AI translators empower new speakers of Lemko to read in their heritage language immediately, without undergoing the thousands of hours of instruction and immersion normally required,” said Petro Orynycz, the linguist and engineer behind the project.
The language revitalization „Manhattan project” is just ramping up, with sights set on delivering translations that are „more human than human”, that is, able to deliver real-time translations more accurate than those of professional field linguists.
The Lemko-English transliteration and translation service has been freely available to the public since 2017 at www.lemkotran.com