|
Newsletter #11 – March 2022
|
|
|
|
Dear reader,
Do you feel like your languages are supported well in the digital world? This is the central question of a short 3-minute survey, in which we want to hear your opinion about the needs of digital support for European languages: Which technologies are already helpful, which are missing? What developments would you like to see in the future? Make your voice heard and participate now. Please also send the survey to your own colleagues and spread it through your own networks. We’re trying to get an unprecedented number of responses, so every single one counts.
At this year’s 13th LREC conference in June, colleagues from ELE organise the workshop Towards Digital Language Equality (TDLE 2022), for which papers can be submitted until 11 April. You can find all the details in the ELE section below.
The partners of the European Language Grid project have been working on updating the ELG user interface and, more importantly, making available translation services for Ukrainian. Three new translation systems from/to Ukrainian have been published by ELG partners in the past two weeks, all with their own set of advantages. The ELG section of the newsletter provides further information on this.
Finally, don’t miss out on the profile of the Association of European Research Libraries (LIBER) at the end of the newsletter and learn why the European Language Equality project “gives LIBER the opportunity to develop knowledge and raise awareness on language technology within the library community”.
With best regards
Georg Rehm
|
|
|
|
Three minutes or less - that's how long it takes to participate our survey on technological support for digital languages in Europe. Click on the image and make your voice heard!
|
|
|
Language Technology and NLP in the news
|
|
|
- “DeepMind and LinkedIn Co-Founders Unveil New Conversational AI Startup Inflection AI” – Voicebot.ai, 11 March 2022
- “DeepMind Researchers Develop Ithaca: A Deep Neural Network For Restoring and Attributing Ancient Texts” – Marktechpost, 14 March 2022
- “Exclusive: Ukraine has started using Clearview AI’s facial recognition during war” – Reuters, 14 March 2022
- “Researchers From Italy Use Machine Learning To Distinguish Different Stages And Severity Of Parkinson’s Disease By Voice” – Marktechpost, 14 March 2022
- “Why Ukraine war misinformation is so hard to police” – CNN Business, 15 March 2022
- “Eastern European IT specialists caught in the crossfire of war” – DW, 16 March 2022
- “Deepfake Zelenskyy surrender video is the 'first intentionally used' in Ukraine war Access to the comments” – euronews, 16 March 2022
- “Open-Source Natural Language Processing: EleutherAI’s GPT-J” – Oodaloop, 16 March 2022
- “A new Stanford study suggests AI still has a bias problem” – Fast Company, 16 March 2022
- “Deepfakes v pre-bunking: is Russia losing the infowar?” – The Guardian, 19 March 2022
- “Encouraging Spanish/Portuguese Collaboration” – language magazine, 19 March 2022
- “Why 2022 is only the beginning for AI regulation” – VentureBeat, 21 March 2022
- “Researchers From China Propose A New Pre-trained Language Model Called ‘PERT’ For Natural Language Understanding (NLU)” – Marktechpost.com, 22 March 2022
- “Russia Is Facing a Tech Worker Exodus” – Wired, 23 March 2022
- “EU rules must focus on AI benefits not curbs, lawmaker says” – Reuters, 23 March 2022
- “NVIDIA brings Alchemab into the Cambridge-1 supercomputer project” – Business Weekly, 23 March 2022
- “Boostlingo Buys VoiceBoxer and Interpreter Intelligence, Adds RSI, Scheduling Tech” – Slator, 23 March 2022
- “Technology-enabled language specialist acquires Dutch cloud data company for £14.5m” – Bdaily News, 23 March 2022
- “How tech volunteers helped build out Ukrainian refugee centers” – Fast Company, 23 March 2022
- “Microsoft to upgrade language translator with new class of AI model” – The Register, 23 March 2022
- “MEPs call for more EU investment in artificial intelligence research” – Research Professional News, 23 March 2022
|
|
- Last week, the European Commission has initiated the ERA4Ukraine portal – a one-stop shop with information for researchers from Ukraine fleeing the war and arriving in Europe.
- ELE partner LIBER, whose profile can be found further below, proves its experience in modern science communication by turning its report for the European Language Equality project into an informational and entertaining Twitter thread.
- Accidental wordplays with the names of countries and languages are always great – Germans becoming germs and Brits turning into bits. But when it comes to the language of Finland, we draw a finnish line.
|
|
|
Over the past two weeks, the European Language Grid initiative has continued providing support for translation between Ukrainian and other European languages. The Institute of Formal and Applied Linguistics of Charles University in Prague have made their Machine Translation service for Ukrainian and Czech available through a free and easy-to-use interface. Further information on the project for “Helping the Ukrainian crisis with language processing tools” can be found on the CUNI website.
A total of 13 languages are available in the translation service of ELG partner Tilde, whose interface is publicly accessible free of charge. The tool can be used for translating documents and entire websites. Furthermore, Tilde has developed a chatbot for the support of Ukrainian refugees in Latvia: The bot helps finding information about support options and informs the local population on ways to help Ukrainians.
Opus-MT (and the University of Helsinki) have also created an interface for their MT services translating text between ten languages, including Ukrainian, Bulgarian and Estonian. The software can be downloaded and run locally on several different systems, keeping all text private and not relying on cloud services.
On to a different subject, the European Language Grid itself: The team at the Institute for Language and Speech Processing (ILSP) in Athens has added several new features to the ELG user interface. Now, tool and resource pages include the number of views, the option for direct citation and ways of sharing resources through various channels. As an example, have a look at the recently added Deaf Reading Assistant by Sign Time GmbH to check out the new functions.
|
|
Selected new tools and resources on the
European Language Grid
|
|
|
|
Czech image captioning, machine translation, sentiment analysis and summarization (Neural Monkey models) – This tool contains trained end-to-end models for the Neural Monkey toolkit for Czech and English, solving four NLP tasks: machine translation, image captioning, sentiment analysis, and summarization. The models are trained on standard datasets and achieve state-of-the-art or near state-of-the-art performance. The same models can also be invoked via the online demo. The Czech models were trained using the SumeCzech dataset, the English models were trained using the CNN-Daily Mail corpus using the standard recurrent sequence-to-sequence architecture. There are several separate ZIP archives here, each containing one model solving one of the tasks for one language. The tool was added by the Institute of Formal and Applied Linguistics of the Charles University on 20 March 2022.
|
|
|
Your help is needed: The European Language Equality consortium is calling all European citizens to make their voices heard and participate in a 3-minute survey on the digital support of their language, in their language: 35 versions of the survey covering most European languages can be found on the ELE website. What do you think? How well are your languages digitally supported? In what areas would better technology support help your languages? And what are your thoughts on Digital Language Equality and the state of European Language Technology?
Participate now – it takes less than three minutes, every response helps! As does sharing the survey with your networks and your colleagues: Please pass on the link and support this effort for Digital Language Equality on terms that represent the opinions and expectations of Europe’s citizens.
To make use of the full potential of Language Technology, progress towards multilingual, efficient, accurate, explainable, ethical, fair and unbiased language understanding is necessary. For the past year, the ELE project has been working on this subject and is now organising the LREC 2022 workshop Towards Digital Language Equality (TDLE 2022) on 20 June in Marseille, France. The organisers of the workshop invite researchers to submit contributions on a range of DLE-related areas such as use cases for LT-deployment, monitoring tools, policy analysis and the societal impact of Digital Language Equality. Papers can be submitted until 11 April 2022. Further information can be found on our ELE blog.
|
|
|
|
Call for Papers: Towards Digital Language Equality workshop at LREC 2022
|
|
|
The ELE consortium – Partner presentation
|
|
|
|
LIBER
LIBER is an association of European Research Libraries and with over 400 members, it is the biggest network of its kind. Through this network, LIBER supports and advocates for research libraries needs and interests. The organisation is a leading voice in Europe on copyright, research data management (RDM) and Open Science and holds a position on the European Open Science policy platform. In addition, LIBER champions European cultural heritage, improving access to collections in European research libraries and extending access to efficient and trusted information sources.
LIBER has strong visibility at the European political level, in the European Parliament and in the European Commission. Moreover, our work is executed thanks to hundreds of volunteers who serve on our Executive Board, Steering Committees, Working Groups and our journal LIBER Quarterly as well as through our participation in international projects.
LIBER in the ELE project:
Within the ELE project, LIBER utilised its network to better understand the landscape of language technology users and consumers at European research libraries. Through surveys, interviews and workshops, we connected with our members to gather data on this important topic. These efforts ensured we publish D2.10, LIBER’s main output from the project. Compiling this information and producing the report furthered both LIBER's understanding of European research libraries and contributed to the ELE project’s goals.
Andrej Vrcon, Head of International Projects:
"The ELE project gives LIBER the opportunity to develop knowledge and raise awareness on language technology within the library community and highlight the role this could play in the future. We see this as an important topic for open science and scholarly communication that should be on the agenda for research libraries."
|
|
|
The next ELT newsletter will be sent out on 12 April 2022. Until then, follow our ELT social media accounts (as linked below) for the latest news!
|
|
|
|
|
