HAITrans represented at CIOL Translators Day

On 15 March 2024, Claudia Wiesinger gave a talk at the Chartered Institute of Linguists’ (CIOL) Translators Day on behalf of the HAITrans research group. Her talk covered the theory of post-editing, the differences between translation, revision and post-editing, and common machine translation errors to watch out for. The lively Q&A showed that the use of machine translation and, more recently, generative AI is a topic of interest among translators (but also increasingly among interpreters). Addressing some common misconceptions about machine translation, Claudia highlighted the importance of understanding how the technology works, so that it can be used to one’s benefit.

In April, the HAITrans team will deliver two workshops on post-editing skills in collaboration with the CIOL, delving deeper into this important topic.


Doctoral project featured in UniVie IMPACT Award series

In July 2023, HAITrans member Claudia Wiesinger was awarded the 2023 IMPACT Award, a prize given to outstanding dissertation projects with the potential to reach target groups outside the scientific community and to achieve social, cultural or economic added value. As part of the award, Claudia’s doctoral project was featured in the UniVie IMPACT Award series. Her research explores ways to alleviate risks associated with the use of machine translation in disaster relief and aims to increase resilience to future disasters by promoting access to translated information. You can find out more about Claudia’s project in an in-depth portrait, as well as in an animated video.


HAITrans present at the 2024 JIAMCATT Annual Meeting

Alina Secară and Dragoş Ciobanu took part in the 2024 JIAMCATT Annual Meeting - (R)evolution? AI’s place within the language professions which took place at UNESCO, 6-8 March, 2024. JIAMCATT is the International Annual Meeting on Computer-Assisted Translation and Terminology, a task force of IAMLADP the international forum and network of managers of international organizations employing conference and language service providers. JIAMCATT membership includes most international organizations, as well as various national institutions and academic bodies, active in the field of terminology and translation. It provides its participants with a forum for debate, exchange of expertise and cooperation in the fields of computer-assisted terminology and translation, interpretation and documentation retrieval.  The meeting included prsentations, demos and workshops on AI use cases and adoption in human translation workflows, AI and green computing, creation of text summaries from videos, open-source workflows for for subtitling, semantic integration and AI drafting and editing tools. Alina and Dragoş shared with the other participants results and plans for their recent projects - Imminent The impact of Speech Synthesis on cognitive load, productivity and quality during post-editing machine translation (PEMT) and the LT-LiDER: Language and Translation: Literacy in Digital Environments and Resources.

2024 JIAMCATT Annual Meeting Group Photo

2024 JIAMCATT Annual Meeting Group Photo


HAITrans member visits Dublin City University (DCU)

HAITrans member Claudia Wiesinger made use of an Erasmus+ staff training mobility to visit Dublin City University between 12 and 16 February 2024. The objectives of this visit were research training, knowledge exchange, and networking. Dublin City University is one of the main research hubs for crisis translation. As such, Claudia’s research interests are closely aligned with those of the researchers at DCU. During her time in Dublin, Claudia met with several seasoned and budding researchers working in the field of crisis translation. She also presented at the School of Applied Language and Intercultural Studies (SALIS) research seminar series. In her presentation titled The power of speech: using speech technologies in translation, revision, and post-editing machine translation (PEMT), Claudia shared an overview of the HAITrans research group’s investigations into speech-enabled working modes. 


HAITrans present at Translating and the Computer 45

On behalf of the HAITrans team, Justus Brockmann presented at the 45th Translating and the Computer conference on 22 November, 2023. The conference was organised by AsLing at the European Convention Centre in Luxembourg and addressed current topics at the interface of translation and technology. The HAITrans contribution presented an investigation into The use of speech technologies and machine translation in institutional translation practices, summarising the findings of a survey answered by 93 professionals involved in translation work at international, national and regional institutions.
A majority of the respondents use one flavour or another of machine translation (MT) in their work and most of them think that the system they are using delivers good or excellent quality. The responses revealed some controversy surrounding the term post-editing, warranting further research into how this practice is viewed and defined in professional practice. Adoption of speech technologies (i.e., using speech-to-text dictation or text-to-speech synthesis) was modest among respondents, but a quarter read aloud the texts they are working on. MT and speech technologies rank highest on the list of training and implementation priorities mentioned by respondents.
A conference paper discussing these and other findings has been submitted for peer review.


HAITrans present at IPCITI 2023

On 24 November 2023, HAITrans team member Claudia Wiesinger presented at the annual International Postgraduate Conference in Translation and Interpreting (IPCITI) held on 23 and 24 November 2023 at Heriot-Watt University in Edinburgh. In her presentation titled “Mixed methods research into crisis translation: Combining surveys, interviews, and eye tracking”, Claudia discussed the suitability of mixed methods research for the context of crisis translation and showed preliminary results from her doctoral project.


Alina Secară was a speaker at the TranslatingEurope Workshop - Now her presentation is available along the other participants on YouTube

Wer nicht persönlich am TranslatingEurope Workshop zum Thema Berufseinstieg und Jobchancen im Bereich Barrierefreiheit am 26.09.2023 im Haus der EU in Wien teilnehmen konnte, hat ab Dienstag, 17.10. 10 Uhr die Möglichkeit, den Workshop auf dem Youtube-Kanal des Übersetzungsdienstes der EU-Kommission anzusehen: https://youtu.be/q5YLxN8ItH0

 

Der Bereich Barrierefreiheit kann für angehende wie auch etablierte Übersetzer*innen und Dolmetscher*innen ein interessantes Betätigungsfeld bieten. In drei Vorträgen geben Alina Secară, Theodoros Sakalidis und Verena Brinda Einblick in die technologischen Möglichkeiten und Anforderungen an barrierefreie Kommunikation, in den Berufsalltag beim Schriftdolmetschen und Live-Untertiteln sowie in das Übersetzen in Einfache und Leichte Sprache. Die Veranstaltung wurde von der Generaldirektion Übersetzung gemeinsam mit dem Österreichischen Verband für Schriftdolmetschen ÖSDV und der österreichischen Translationsplattform organisiert und fand zur Feier des Internationalen Tags des Übersetzens der FIT (Fédération Internationale des Traducteurs) und des Europäischen Tags der Sprachen im Haus der EU in Wien statt.

 


Two HAITrans projects win Impact Award 2023

HAITrans members Claudia Wiesinger and Raluca Chereji won the 2023 Impact Award for their doctoral projects, marking the second consecutive year that two out of the eight prizes are awarded to researchers within the Centre for Translation Studies. The prize is offered by the University of Vienna and funded by the City of Vienna Cultural Affairs in recognition of outstanding doctoral projects with the potential to achieve tangible social, cultural or economic impact beyond academia. Claudia and Raluca received their award at the end of a pitching event judged by a panel of experts in communication, entrepreneurship and funding. Claudia presented on potential applications of speech-enabled machine translation post-editing (PEMT) in crisis translation, while Raluca discussed ways of improving doctor-patient communication in translation using linguistic analyses and translation technology. As part of this award, both will receive tailor-made video productions presenting the impact of their doctoral research, and will benefit from promotion on the University of Vienna media channels.

More information can be found on the Impact Award website: https://forschung.univie.ac.at/en/services/funding/early-career-researchers/impactaward/


HAITrans contribute to TranslateCluj 2023

On 24 March 2023, HAITrans team member Raluca Chereji presented at the annual TranslateCluj conference held between 23-25 March 2023 in Cluj-Napoca, Romania. In her presentation titled “Researching medical translation: Challenges, practices and technologies”, Raluca outlined new trends in the translation industry, such as increased technologisation and the rise of the healthcare sector, before discussing methods and methodologies in translation research, with a focus on the medical domain. She also shared preliminary results from her doctoral research, which includes a corpus linguistics analysis of clinical trial Informed Consent Forms in Romanian and an international questionnaire for professional medical translators.


HAITrans present at Smart Diaspora 2023 Conference

The HAITrans team delivered three presentations as part of the Smart Diaspora 2023 Conference held between 10-13 April 2023 in Timişoara, Romania and hosted by the West University of Timişoara. Univ.-Prof.Dragoş Ciobanu, Senior Scientist Dr. Alina Secară and PhD student Raluca Chereji contributed to the workshop series titled Traducere şi comunicare interculturală – de la digitalizate la inteligenţa artificială. Noi direcţii în predare, cercetare şi practică”.

On 11 April, Prof. Ciobanu delivered the workshop’s first plenary talk titled “Traducerea automată neuronală şi traducerea asistată de calculator în practică şi cercetare”, which looked at how Neural Machine Translation and Computer-Assisted Translation tools are being used in the industry, as well as in academic research. This was followed by a presentation by PhD student and University Assistant Raluca Chereji on “Recunoaşterea vocală automata în traducerea şi post-editarea textelor din domeniul medical”. Raluca outlined some of the main advantages and disadvantages of using Automatic Speech Recognition within translation and post-editing workflows, and demonstrated a potential use case for translations of medical texts aimed at patients.

On 12 April, Dr. Alina Secară delivered a second plenary talk titled “Tehnologiile de subtitrare pentru localizarea şi accessibilizarea conţinutului media”. Dr Secară’s presentation combined an overview of subtitling technologies and use cases with findings from research conducted at the University of Vienna and the University of Leeds.

The sessions concluded with a joint HAITrans workshop on “Noi tehnologii pentru cercetarea procesului de traducere: eye-tracking”, during which the team discussed some of the current HAITrans research projects which use eye-tracking of study participants.

Further information about the Smart Diaspora 2023 Conference is available here:

 

https://www.diaspora-stiintifica.ro/

 

 

 


HAITrans invited to present at the UN InnoVent 2

UN Photo / Matija Potocnik UN InnoVent 2 participants

UN Photo / Matija Potocnik UN InnoVent 2

UN Photo / Matija Potocnik UN InnoVent 2 online speakers

UN Photo / Matija Potocnik UN InnoVent 2

The ZTW Univ.-Prof. Dragoş Ciobanu and Dr. Alina Secară were invited to speak at the InnoVent 2 - Innovation in Conferencing event organised by the United Nations (UN) in Geneva on the 30th of March, 2023. In the "Speech-enabled Post-editing of Machine Translation" and the "Speech Recognition for Conferencing" sections they were joined by colleagues from UN Geneva, the European Commission, the World Intellectual Property Organisation (WIPO) and the International Labour Organisation (ILO) to discuss the latest in terms of development, implementation, and use of speech technologies in the context of multilingual communication and international conferencing.

Further information about the event can be accessed on the dedicated InnoVent 2 website.


HAITrans contribute to GALA 2023 Dublin

On 13 March 2023, HAITrans member Claudia Wiesinger presented on behalf of the research group at the annual Globalization and Localization Association (GALA) conference in Dublin, Ireland. In the session titled “Speech-enabled MTPE: Effective or Distracting?”, Claudia showed preliminary results of a research project funded by Imminent, the research arm of the company Translated. In the project, eye tracking and other methods are used to measure the effectiveness and cognitive impact of listening to speech synthesis (text-to-speech) during machine translation post-editing.

 

The call for participation is still open:

 

If any more professional translators based in or around Vienna wish to contribute to this research project, we look forward to hearing from you. We are looking for participants with at least 3 years of translation experience and, ideally, at least 1 year of post-editing experience. Participants should be working from English into German and have German as a first language. The total duration of the experiment will be up to three hours, and you will be paid 100 EUR for your participation. To register your interest, please fill in our recruitment questionnaire in English (https://survey.trans.univie.ac.at/index.php/991875?lang=en) or in German (https://survey.trans.univie.ac.at/index.php/991875?lang=de). We will contact you if your profile matches our experiment requirements.

 


Call for participation!

We are looking for professional translators based in or around Vienna, Austria, working from English into German who have German as a first language to contribute from the second half of January, 2023, to the research project “The impact of Speech Synthesis on cognitive load, productivity and quality during post-editing machine translation (PEMT)”. The project is led by Univ.-Prof. Dragoș Ciobanu and funded by Imminent (the research arm of Translated).

HAITrans is committed to researching novel uses of new and existing technologies which help translators adapt to the changing nature of the translation market, deliver high-quality translations, and improve productivity. The current research project will investigate the impact which speech synthesis (i.e. hearing an artificial, computer-generated voice reading the source and target segments) has on translators who post-edit machine translation output from English into German.

Your participation will consist of answering one pre-experiment and two experiment online questionnaires, as well as of taking part in an eye-tracking study in our Vienna-based lab where your eye movements on a computer screen, the computer sound, as well as your keyboard presses and mouse clicks will be recorded. Your face and voice will not be recorded. Your data will be fully anonymised within one month of your participation, so it will not be possible to trace it back to you.

Your task in this experiment will be to post-edit four short documents (factsheets about COVID-19 with a combined total number of words of 1,423 for the English original) in the Matecat CAT tool project, at times hearing the source and target segments being spoken out, and at other times without hearing any such synthetic sound.

An eye tracker will be used to record your gaze during the experiment. The computer screen and computer interactions you will have during the experiment will be recorded for later annotation and anonymous comparison with other experiment participants. Your details will be anonymised and you will be assigned a random experiment ID, so it will not be possible to trace back your performance to you personally. The effectiveness of your post-edits will be evaluated separately and will also be anonymised.

The total duration of the experiment will be up to three hours and you will be paid 100 EUR for your participation. You will need to fill in a University of Vienna freelancer datasheet and invoice the University of Vienna Centre for Translation Studies. You will be responsible for paying any tax associated with this amount as part of your earnings.

To register your interest, please fill in our recruitment questionnaire in English or in German. We will contact you shortly afterwards if your profile matches our experiment requirements.


Linguists from the European Central Bank at ZTW

On 28 November 2022, three members of the Language Service at the European Central Bank (ECB) visited the ZTW. In the first part of the session, Pauline Pitcher, Markku Haverinen, and Peter Wilkinson explained to students what it is like to work as a translator or editor in the ECB’s Language Service, covering the typical profile of its linguists, the varied day-to-day tasks and texts they work on, their focus on editing, how they learn on the job and the tools of the trade. They also provided information about traineeships, as well as tips on how to apply and prepare for the selection process.

The second part of the session consisted of an English editing workshop, during which students could see first hand what editing for the ECB entails. The presenters discussed the different types of errors editors need to be mindful of, as well as other important aspects such as clear writing, foreign language interference and editing for social media. Students were also able to try their hand at editing in English during the many practical exercises and examples presented throughout the workshop.


HAITrans present at the 44th Translating and the Computer conference

The HAITrans team was involved in three presentations at the 44th Translating and the Computer conference hosted by the International Association for Advancement in Language Technology (AsLing) in Luxembourg on 24-25 November 2022.

On 24 November, HAITrans members Miguel Rios and Raluca Chereji presented on a project conducted together with Prof. Dragoș Ciobanu and ZTW Senior Scientist Alina Secară. In their presentation titled "Impact of Domain-Adapted Multilingual Neural Machine Translation in the Medical Domain", Miguel and Raluca outlined the quality improvements brought about by fine-tuning a Multilingual Neural Machine Translation (MNMT) model with in-domain data for the medical domain, compared with an out-of-domain MNMT model, for the English-Romanian language pair. Miguel and Raluca used automatic metrics and human error annotation based on a terminology-specific error typology to show that the in-domain MNMT model outperformed the out-of-domain MNMT model across most metrics, producing fewer terminology errors across most error categories surveyed.

On 25 November, in the morning, Raluca Chereji and Justus Brockmann, on behalf of the entire research group, presented on “The use of speech technologies in translation, revision, and post-editing machine translation (PEMT)”, providing an overview of HAITrans research projects that investigate applications of automatic speech recognition and automatic speech synthesis. The results of these investigations indicate that speech synthesis can support revision and PEMT, especially in terms of improved attention to errors. Future efforts include investigating the impact of speech synthesis on post-editors’ cognitive load, among other things. Three doctoral research projects within the HAITrans group will further investigate applications of speech recognition and synthesis in different scenarios.

On the same day, Dragoș Ciobanu chaired a panel discussion on the “Past, present and future of speech technologies in translation”. Dragoș was joined by panelists Julián Zapata, Carlos Teixeira, Marcin Feder, and Alina Secară. The panelists discussed how speech recognition and synthesis technologies have reached levels of maturity that have recently boosted their prominence. Examples in the area of translation include experimental translation and PEMT set-ups that rely on touch and voice rather than mouse and keyboard, new university courses that integrate the use of speech recognition in translator training, or a real-time speech-to-text plus machine translation solution being deployed to increase accessibility at the European Parliament. Accessibility was identified as another major driver behind the increasing prominence of speech technology, with new legal accessibility requirements and the large volume of content they affect warranting interest in the use of automated speech-to-text and text-to-speech solutions.


HAITrans workshop for TCLoc students

On 18 November 2022, HAITrans member Claudia Wiesinger ran a workshop on speech technologies and their use in translation, revision and post-editing machine translation (PEMT) for students of the Master’s in Technical Communication and Localization (TCLoc) at the University of Strasbourg.

In the first part of the workshop, Claudia discussed the potential benefits of automatic speech recognition and automatic speech synthesis for translators’ work, and presented current research on their qualitative, behavioural, and cognitive impact. The second part of the workshop was a hands-on session during which the students worked on practical translation and PEMT tasks using a combination of Microsoft Word and the free, web-based CAT tool Matecat.


HAITrans at 1st International Conference of Translation, Graphic Medicine and Doctor-Patient Communication

HAITrans member Raluca Chereji presented at the 1st edition of the International Conference of Translation, Graphic Medicine and Doctor-Patient Communication, held in hybrid format on 3-4 November 2022 and organised by the University of Córdoba, Spain. In her presentation titled “Investigating the language of patient information: a corpus linguistics analysis on the readability and lay-friendliness of Romanian clinical trial Informed Consent Forms”, Raluca shared the results of an analysis of 20 consent forms in Romanian. Raluca focused on measurements of readability, as well as lexico-syntactic features including part-of-speech classification, expert terminology and linguistic interference, to highlight potential areas for improvement in Romanian patient-facing translations.

 

 


HAITrans presents at DAML Conference 2022

HAITrans member Raluca Chereji presented at the 20th annual Translation and Interpretation, Past, Present and Future conference organised by the Babeş-Bolyai University Department of Applied Modern Languages and Centre for Language Industries of the Faculty of Letters, and which took place in Cluj-Napoca, Romania. Raluca presented on “The Challenges and Opportunities of Using Automatic Speech Recognition and Machine Translation Post-Editing (PEMT) in Romanian Translations of Expert-to-Lay Communication in the Medical Domain”. The presentation, which was based on Raluca’s ongoing doctoral project, combined an overview of current developments in ASR with examples of the expert-to-lay bias present in patient-facing translations, and included a demonstration of what integrating ASR into Romanian medical translations could entail from a technical and practical standpoint.


Machine Translation Training for UN Language Professionals

Azucena Bajo, French Translation and Text Processing Chief UNOV and the ZTW Training Team

Azucena Bajo, French Translation and Text Processing Chief UNOV and the ZTW Training Team

On the 27th of October, 2022 Prof. Dragoș Ciobanu, together with the HAITrans Team and ZTW Senior Scientist Alina Secară provided a one-day training titled "Translation in the Era of Machine Translation and Post-Editing" at the UN Vienna International Centre. Over 50 in-house senior and junior staff from the translation services from UN Vienna (UNOV), UN Geneva (UNOG), and UN New York (UNHQ), as well as representatives from the IAEA translation services, took part in the event, face-to-face or online.

With concrete examples of neural machine translation (NMT) use from the industry, institutions and training contexts, as well as data from research, the presenters discussed the pros and cons of neural machine translation in professional translation contexts. The topics addressed ranged from scenarios suitable for post editing of machine translation (PEMT), quality and productivity in PEMT workflows, and differences between PEMT and revision, to strategies for PEMT, integration of further tools in PEMT, and impact of MT on linguists, translation processes, and products. The aim was to facilitate fruitful interactions between participants and stimulate discussions about technological advancements and their impact on the occupational status, visibility, and work of professional translators.

The event was part of the collaboration between the University of Vienna and UNOV within the IAMLADP Universities Contact Group. The Universities Contact Group of the IAMLADP Working Group on Training acts as liaison between international organizations and universities as training providers, and promotes information exchange, cooperation and better relations between the two constituencies.

Presentation at the European Conference of Speech-to-Text Interpreters

Alina Secară delivers her presentation at ECOS

Alina Secară delivers her presentation at ©ECOS

Alina Secară delivered the presentation titled Captioning and speech to text interpreting for the undeclared audience – some knowns and unknowns at the European Conference of Speech-to-Text Interpreters (ECOS) Vienna 2022: PostPandemicPerspectives which too place between 26 and 28 of August 2022 in Vienna. 

The international conferece was organised by the Centre for Translation Studies at the University of Vienna (ZTW) in cooperation with the Board of ÖSDV – Austrian Association of Speech-to-text Interpreters and included several contributions from the ZTW staff and former students. 

There is wide agreement in the broadcasting world that captioning, traditionally an access service created primarily for audiences with hearing loss, is consumed by a much wider audience than that for which the service is originally intended for. Captioning can be prepared in advance (in such cases the term pre-prepared intralingual subtitling is also used), delivered live (usually using speech recognition technology in a process called respeaking when used in broadcasting, or speech to text interpreting particularly in live face-to-face settings) or it can be hybrid, prepared in advance and synchronised live (frequently in theatre settings). Alina's presentation discussed if and how captioning, irrespective of methods or settings in which is used, can benefit people outside the classic accessibility context. To do this, examples from two separate spheres were introduced: first, two theatre captioning projects the presenter took part in as an accredited Stagetext theatre captioner working with a British theatre; secondly, an educational project where the presence of live captions produced via speech recognition for MA lectures went beyond the provision of access for one deaf student to benefit a much wider variety of students.

For a summary of the conference, in German, click here.


HAITrans wins Imminent research grant

Univ.-Prof. Dragoș Ciobanu was awarded one of the five Imminent 2022 grants to carry out the reasearch project investigating "The impact of Speech Synthesis on cognitive load and productivity during post-editing machine translation (PEMT)"

In the project proposal, Prof. Ciobanu wrote: "The increased fluency of neural machine translation (NMT) output recorded in certain language pairs and domains justifies its large-scale deployment, yet professional translators are still cautious about adopting this technology. Among their main concerns is the already-documented “NMT fluency trap” that causes translators to miss significant Accuracy errors masked by the NMT output’s high fluency.

Our UniVie HAITrans research group has been investigating the potential of speech technologies – synthesis and recognition – to improve the quality of professional and trainee translators’ work. This project will specifically build on our experiments involving speech synthesis in the revision and post-editing processes which show a superior level of Accuracy error detection and correction when synthesis is present. Given our findings that revising with speech synthesis does not impact negatively on the revisers’ cognitive load, we will use our eye-tracking lab to investigate cognitive load and productivity when post-editing with sound versus in silence.

Should our PEMT findings mirror our work on revision, we expect translators to feel reassured that, when integrating speech synthesis into their PEMT workflows, this technology will help them avoid the NMT fluency trap."

Prof. Dragoș Ciobanu was awarded one of the five Imminent 2022 research grants


10th EST Congress Presentation - Media Accessibility Training in the EMT Network

EST Conference Group Photo

EST Conference Group Photo

Alina Secară delivered the Media Accessibility Training in the EMT Network presentation part of the Accessibility in Context: Inclusiveness in Specialised Translation and Interpreting Panel at the 10th Congress of the European Society for Translation Studies (EST) which took place in Oslo between the 22nd and 25th of June, 2022. The presentation, prepared with colleagues from the Leiden University, Constantine the Philosopher University in Nitra and University of Westminster, all members of the European Masters in Translation (EMT) Working Group on Audiovisual Translation (AVT) and Media Accessibility (MA), offered an overview of AVT and MA training practices within the EMT member universities. Quantitative and qualitative data from a 2020 survey containing responses from 55 EMT training institutions was presented. Of the 55 Masters-level programmes that participated in the survey, 43 (78%) reported providing training in audiovisual translation and/or media accessibility. The presentation discussed the type of AVT and MA courses (audio description, subtitling, film studies, web accessibility) included in these Masters-level programmes; their nature (compulsory or elective, dedicated modules or part of a more general module), and the technology supporting this training. These findings were then discussed in the context of the competences required of specialised translators and interpreters, pointing to overlooked areas of training and potential implications for future professionals specializing in AVT and accessibility.


HAITrans invited at TranslateCluj Back 2 Tech 2022

TranslateCluj Group photo

TranslateCluj Group photo

Dragoș Ciobanu and Alina Secară were invited to present at the TranslateCluj Back 2 Tech conference which took place in Cluj, Romania between 19th and the 20th of May, 2022.

To use or not to use speech synthesis for translation, revision or MT post-editing was the title of Dragoș Ciobanu’s presentation to an audience of professional translators and interpreters, as well as academic colleagues present at the 2022 edition of TranslateCluj. His talk offered an overview of recent research showing that both speech recognition and synthesis can positively influence the output quality, language professionals’ productivity, and workspace ergonomics associated with translation, revision, post-editing machine translation (PEMT), audio-visual translation, as well as interpreting processes. He offered examples showcasing current integrations of speech technologies in Computer Assisted Translation/PEMT and offered insights into future implementations of speech technologies that could benefit professional linguists, as well as content creators and consumers.

In her presentation, titled Accessible communication and opportunities for linguists, Alina Secară stressed the need to develop content, services and applications that meet accessibility requirements - be it online, for television broadcast, for cultural events or face-to-face situations - following the European Accessibility Act (2019/882) and the EU Directive on the Accessibility of Websites and Mobile Applications (2016/2102). Throughout Europe the implementation and delivery of access services is far from uniform and, in some countries, still in their infancy. As national legislation regarding the provision of access services is gradually passed, demand for specialists in this area is likely to rise. This presentation therefore highlighted opportunities likely to arise from these legislative transformations and requirements, and provided an overview of such services supported by examples of existing implementations in various European countries and contexts.


Translating Europe Workshop: Cross-border lessons in European audiovisual translation

Translating Europe Workshop AVT Panel

Translating Europe Workshop AVT Panel

Alina Secară was an invited to be a speaker in the Current professional profile of audiovisual translators panel  at the Translating Europe Workshop (TEW) Cross-border lessons in European audiovisual translation. The TEW event, which took place on June 10th, 2022 at the House of European Union in Vienna, brought together experts in audiovisual translation from seven European countries. Trainers and practitioners discussed legal aspects and in particular author’s rights for audiovisual translators, exchanged infomration on market requirements, trends and training opportunities and presented current practices in subtitling, dubbing, and voice-over from various countries. Alina was joined on the pannel by ZTW colleague Birgit Grübl (who also represented the public broadcaster ORF, Austria), Agnieszka Szarkowska (University of Warsaw, Poland), Emília Perez (Constantine the Philosopher University, Slovakia), Mandana Taban (Taban Translating Films, Austria) and was moderated by Amalie Foss (AVT-E, Denmark). The recording of the full event can be accessed here. 


ZOO Digital – ZTW Academic Partnership in its Second Year

ZOO Digital Academy Partner

ZOO Digital Academy Partner

As we enter our second year of Academic Licence Agreement with ZOO Digital and its cloud-based subtitling platform ZOOsubs we look back at the activities this collaboration has enabled.

Following the signing of an Academic Licence Agreement in April 2021, the University of Vienna Centre for Translation Studies has been successfully integrating the cloud-based ZOOsubs platform in its MA in Translation courses. Students enrolled in courses focussing on media localisation and accessibility have been carrying out tasks such as captioning, subtitling and audio description using ZOOsubs either in individual tasks or as part of collaborative team projects. In addition to familiarising themselves with typical subtitling workflows – Transcription, Spotting, Technical QA, Translation and Client Review – the platform has been enabling students to collaborate in a professional environment in revision or post-editing tasks, and employ translation quality assessment principles using the ZOO Issues typology and Comments functionalities. Therefore, access to ZOOsubs supported them in developing both creative, translation-related skills, as well as technical proficiency. Students enrolled in an Accessibility module were also able to use the platform to create collaboratively audio descriptions. Having access to the cloud-based ZOOsubs platform made the transition to online teaching during the pandemic particularly stress-free and offered both students and teachers a professional environment which could be easily accessed.

Alina Secară who implemented the agreement and firmly embeded ZOOsubs in her teaching at the ZTW, also collaborated with ZOO Digital professionals to co-present or run teacher training and student training sessions at various events, thus enabling colleagues and students from other universities to access the platform. Some of these events were the Übersetzen und Technologien Workshop organized by the European Commission representation in Austria, the University of Vienna, the University of Graz, the University of Innsbruck and UNIVERSITAS Austria on Hieronymus Day in 2021, The Translating Europe Workshop From Translation to Accessibility EMT Train the Trainer Summer School and the Workshop on Subtitling at the Nitra Summer School on Audiovisual Translation: Translation, Technology and Inclusion Summer School both held in June 2021.

We would like to thank ZOO Digital and we look forward to continuing our collaboration!


HAITrans contribute to NeTTT 2022

04.–06.07.2022

HAITrans members Claudia Wiesinger, Raluca Chereji, and Justus Brockmann attended the New Trends in Translation and Technology (NeTTT) conference which took place in Rhodes, Greece. Justus presented on "Using error annotation to support post-editing in the context of translator training", discussing findings from a HAITrans study that introduced speech synthesis technology in an error annotation and post-editing task carried out by students at the University of Vienna Centre for Translation Studies.

An analysis of the annotations performed by the students, as well as their answers to the study questionnaires, showed that there was considerable variation in students' annotation behaviour and preferences, and that error annotation was a challenging task for them. Yet, a majority perceived the task to be useful and the average distribution of error types in their annotations was to some extent comparable to the distribution in the Gold Standard produced by the research team. With this in mind, the presentation concluded encouraging an emphasis on Translation Quality Assessment practices - especially typology-based error annotation - in translator training to empower students to confidently deal with the strengths and weaknesses of MT in their future roles and train their ability to provide structured feedback to MT stakeholders.


HAITrans contribute to EAMT 2022

01.–03.06.2022

HAITrans members Claudia Wiesinger and Dragoș Ciobanu contributed to this year's 23rd Annual Conference of the European Association for Machine Translation (EAMT), which took place in Ghent, Belgium. The presentation entitled "Error Annotation in Post-Editing Machine Translation: Investigating the Impact of Text-to-Speech Technology" discussed findings from a study that introduced speech synthesis technology in an error annotation and post-editing task carried out by students at the University of Vienna Centre for Translation Studies. Focusing on the error annotation data, the analysis found that participants under-annotated fewer MT errors in the speech synthesis condition compared to the silent condition. At the same time, more over-annotation was recorded. Finally, annotation performance corresponds to participants’ attitudes towards using speech synthesis.

The paper that reports on these findings was published in the conference proceedings.


REVALORISE+

HAITrans university assistant and PhD student Raluca Chereji has been accepted onto the REVALORISE+ international project taking place between October 2022 and January 2023 and organised in conjunction with the University of Vienna. The REVALORISE+ training programme is aimed at Social Sciences and Humanities researchers working to extend the reach of their projects beyond academia. Raluca will take part in the Valorisation Pathway and work on developing her PhD research into an entrepreneurial valorisation plan which achieves a tangible societal impact.

 

More information on the REVALORISE+ project can be found here:

https://revalorise.eu/


Translating Europe Workshop: AVT workflows and the role of automation

The Università degli Studi Internazionali di Roma – UNINT, the Directorate-General for Translation (DGT) of the European Commission (Rome Field Office) in collaboration with the University of Antwerp (Belgium), Kaunas University of Technology (Lithuania), ISCAP (Portugal), Constantine the Philosopher University in Nitra (Slovakia) and the University of Vienna (Austria) are inviting you to take part in the Translating Europe Workshop (TEW) titled AVT workflows and the role of automation, 24 May 9:30- 4:30 p.m. (CET) on Zoom. The event is the second workshop in the From Translation to Accessibility: EMT Train the Trainer Summer School series designed and organized by this consortium of universities and supported by the DGT. The aim of the series is to share expertise and good practice in AVT (audiovisual translation) and accessibility training across Europe, through the involvement of EMT universities, EU institutions, LSP providers and software companies.

What: TEW AVT workflows and the role of automation

When: 24 May 9:30-4:30 p.m. (CET) on Zoom

Where: Zoom, please see the https://www.unint.eu/calendario-eventi/translating-europe-workshop-24-maggio-2022.html for the link

Who: The programme, as well as further details can be accessed here: https://www.unint.eu/calendario-eventi/translating-europe-workshop-24-maggio-2022.html 

We are looking forward to seeing you online!


HAITrans present at IATIS Training Event

On 21 April, the International Association for Translation and Intercultural Studies (IATIS) and the Institute of Translation and Multilingual Communication at TH Köln University of Applied Sciences jointly organised a training event on "Neural Machine Translation - Foundations, Applications and Implications". Researchers and industry experts shared insights into the historical foundations and current developments, industry implementation, post-editing competence, ethics, and education in the context of neural machine translation (NMT). HAITrans members Claudia Wiesinger, Raluca Chereji and Justus Brockmann were invited to present their NMT-related PhD research projects. The three projects investigate applications of speech technologies (automatic speech recognition/automatic speech synthesis) in translation and NMT post-editing.

With 300 registered and more than 100 active participants throughout the day, the event showed that there continues to be great interest in NMT as currently one of the most-discussed technologies in the field of translation.


HAITrans member as recipient of the 2022 GALA Rising Star Scholarship

HAITrans member Claudia Wiesinger is the recipient of the 2022 GALA Rising Star Scholarship, which will enable her to attend the GALA 2022 conference that is taking place in San Diego from 24th to 27th of April.

In order to qualify, Claudia submitted a video application addressing the prompt “How can you adapt teaching/learning methodologies to meet both the students' and the industry's needs?”. In her submission, she proposes an open-access, web-based platform that can be used to practise post-editing and error annotation with real-world examples and explanations from Language Service Providers (LSPs) and translation departments across the globe.

You can watch the video submission and an interview with Claudia about her win here.


Call for abstracts - Conference on teaching translation and interpreting in the age of neural machine translation

HAITrans Prof Dragoș Ciobanu and Alina Secară, members of the Scientific Committee of the Conference on teaching translation and interpreting in the age of neural machine translation (NMT) which will take place on 29 & 30 September 2022 at ULB, Campus du Solbosch Brussels, Belgium, together with the Conference Chair, Prof Pascaline Merten from Université libre de Bruxelles, invite contributions to this conference from colleagues working in the following areas:

- NMT and the translation profession

- Integrating NMT, post-edition and revision into translator and interpreter training

- Developments in corpus-driven machine translation and interpreting

 

We invite researchers, professional translators and interpreters, translator and interpreter trainers to share their experience and ideas in the form of a 20-minute presentation followed by a discussion. If interested, please submit anonymized abstracts of approximately 500 words, including relevant references by 28 March 2022 to pascaline.merten@ulb.be

For further details please see the attached PDF.

 

Important dates:

 

Deadline for abstract proposal (500 words): 28 March 2022

Notification of acceptance: 2 May 2022

Deadline for submission of final extended abstract (2,000 words): 15 August 2022 

Conference: 29 & 30 September 2022, Brussels, Belgium

 


Enhancing Multilingual Machine Translation with Terminology Information for the Medical Domain

01.03.2022

HAITrans member Miguel Angel Rios Gaona successfully applied for a grant to develop machine translation (MT) research for the medical domain from the Google Cloud research credits programme. The grant provides access to computing resources worth 5K USD. In the medical domain the accurate translation of terminology is crucial for exchanging information across international healthcare providers or users. Our goal is to enhance medical MT models based on terminology information.


HAITrans members present at the third international Unlimited! conference

11.02.2022

HAITrans members Alina Secară and Raluca Chereji presented at the third international Unlimited! conference – Innovation for Access: New Interactions – organised by the University of Antwerp and which took place online. In their presentation entitled "Describing personal characteristics in audio introductions – towards equitable lexical choices", Alina and Raluca shared the results of an in-depth corpus analysis of 52 stage play audio introductions in English. They analysed the linguistic characteristics of their corpus in terms of readability, lexical density and lexical sophistication using a variety of metrics. They then focused their investigation on the potential gender bias and imbalances in the rendering of personal characteristics of the male and female characters in their corpus. In particular, they considered descriptions of age (represented using both explicit and implicit wording), social class in conjunction with skills and abilities, personal characteristics and value judgements. Alina and Raluca also provided qualitative evidence for the structural and chronological organisation of audio introduction elements, thus contributing to existing research on their structural variety and consistency.


A warm welcome to the newest addition to the HAITrans research group

15.11.2021

This month, the HAITrans research group welcomed a new member: Miguel Rios, PhD. He will be supporting the group as a Postdoc in Machine Translation. With an interest in developing models for text representation and machine translation, Miguel will be strengthening the group’s expertise in building the internal capacity for conducting data-driven research of machine translation technologies and practices.


ZTW selected to present within the Campus Aktuell Universität Wien initiative

On the 17th of March, 2022 Elisabeth Fraller and Alina Secară will provide an introduction to subtitling and film translation to the general public, as part of the Campus Aktuell initiative at the Universität Wien.

Warum werden in Ländern wie Schweden oder Portugal fast ausschließlich untertitelte Filme in den Kinosälen und im Fernsehen gezeigt, während andere Länder wie Österreich, Deutschland oder Italien als „klassische Synchronisationsländer“ gelten? Was hat das alles mit der historischen Entwicklung dieser Länder und mit Sprachpolitik zu tun? Und welche Auswirkungen auf den Fremdsprachenerwerb sind damit verbunden? Wie und wo arbeiten Untertitler*innen eigentlich, und nach welchen Regeln werden Untertitelungen erstellt? Und welche Rolle spielen dabei Software und Kognitionswissenschaft?

Diese Fragen und noch viel mehr werden in einem spannenden und anschaulichen Vortrag mit praktischen Beispielen und interaktiver Einbindung des Publikums vorgestellt und erläutert.

Vortrag in deutscher und englischer Sprache.

Thursday, 17th March, 2022, 17:00 - 18:30 

Alte Kapelle am Campus der Universität Wien

Vortragende:

Mag.a Elisabeth Fraller, M.A.; Lektorin für literarisches und mediales Übersetzen am Zentrum für Translationswissenschaft der Universität Wien; Filmuntertitlerin im Österreichischen Filmmuseum und Kunsthistorischen Museum Wien.

Alina Secară PhD, MA; Senior Scientist am Zentrum für Translationswissenschaft der Universität Wien.

NOTE: Due to the current COVID-19 situation this event was rescheduled from 22nd of November, 2021. 


HAITrans present at LocWorldWide45

21.10.2021

HAITrans Senior Scientist Alina Secară joined Tabea De Wille from the University of Limerick, Jan Grodecki from the University of Washington, and Peng Wang from EDUinLOC in the Meet the Academics panel, EDUinLOC Track at LocWorldWide 45 - Pushing Bounderies, 19-21 October, 2021. 

Together with industry representatives they discussed how to best bridge the gap and build conversations between academia and industry, how to support students to develop a career in localization, we well as how to adapt to new realities and continue to push boundaries. 


HAITrans present at the EMT Network meeting in Leipzig

21.10.2021

HAITrans Senior Scientist Alina Secară was invited to share her experience on integrating accessibility in the MA translation curriculum with colleagues, students and representatives of the EU Directorate-General for Translation at the EMT Network meeting in Leipzig, Human centred education – human centred translation.

In her presentation Translation meets Accessibility, part of the panel Accessibility in Translation, she highlighted the main aspects linked to accessibility as part of curriculum development and innovation, impact of automation, and interdisciplinary and participatory models of accessibility.


HAITrans members run sessions on speech recognition and subtitling for Hieronymus Day

30.09.2021

On this year’s Hieronymus Day, HAITrans members Dragoș Ciobanu, Claudia Wiesinger and Alina Secară held three parallel sessions as part of a workshop for translation and interpreting students organised by the European Commission representation in Austria, the University of Vienna, the University of Graz, the University of Innsbruck and UNIVERSITAS Austria. The full programme can be found here.

Dragoș Ciobanu and Claudia Wiesinger introduced students to speech recognition using Dragon NaturallySpeaking while Alina Secară and Emilia Perez from UKF Nitra discussed subtitling using ZOO Digital.


HAITrans at 7th IATIS Conference: The Cultural Ecology of Translation

15.09.2021

HAITrans member Claudia Wiesinger participated in the 7th IATIS Conference with the theme “The Cultural Ecology of Translation”, which took place in a hybrid format (online and at the Universitat Pompeu Fabra in Barcelona) from 13th to 17th September 2021. In her presentation titled “Translating, revising and post-editing with automatic speech synthesis: new technologies, new practices?”, she shared some of the findings of a research project which investigated the impact of integrating automatic speech synthesis into a French into English revision task.

This study, which was conducted by HAITrans members Alina Secară and Dragoș Ciobanu, as well as Valentina Ragni, found that compared to their colleagues who revised in silence, revisers who also heard an automatically generated audio version of the source text: identified and corrected more translation Accuracy errors; made statistically significant shorter fixations on the source segments; had a similar attention distribution across the areas of interest investigated; consulted almost the same number of external resources.

More information on the 7th IATIS Conference can be found on the dedicated website.


HAITrans present at the Łódź-ZHAW Duo Colloquium on Translation and Meaning

03.09.2021

HAITrans members Dragoș Ciobanu and Claudia Wiesinger presented research conducted in collaboration with colleagues Alina Secară and Justus Brockmann at the Łódź-ZHAW Duo Colloquium on Translation and Meaning, which took place online from 2 to 3 September 2021. More information on the ZHAW session of the Duo Colloquium can be found here.

In their presentation titled “Speech-enabled machine translation post-editing (PEMT) in the context of translator training”, the two UniVie HAITrans research lab members shared some of the findings of a study which introduced undergraduate and postgraduate students to the task of error annotation, as well as speech-enabled PEMT, a novel practice that involves the use of a computer-generated, synthetic voice that reads aloud the source and/or target text while post-editing.


A warm welcome to two new HAITrans research group members

15.07.2021

This month, the HAITrans research group welcomed two new members: Raluca Chereji, BA MSc and Justus Brockmann, BA MA. They will be supporting the group as graduate research assistants (prae doc) while completing their respective doctoral projects. Both have extensive experience working in the language services industry, which will make them valuable additions to the team.


HAITrans present at the LEAD-ME Eye Tracking Research Summer Training School, 5-9 July, 2021

05.07.2021

HAITrans was present at the LEAD-ME Summer Training School, 5-9 July, 2021, organised online by the SWPS University of Social Sciences and Humanities, Poland and the University of Warsaw, Poland, as part of the Leading Platform for European Citizens, Industries, Academia and Policymakers in Media Accessibility (LEAD-ME) COST Action CA19142.

Senior Scientist Alina Secară joined the summer school organisers as an invited tutor, and Raluca Chereji and Claudia Wiesinger, graduate research assistants (prae doc), took part as students. The summer school focused on eye tracking in media accessibility research and included fundamental concepts of eye tracking research, focusing on a variety of research methods, tools, statistical analyses and research communication. Participants took part in a series of talks and workshops, and they were also offered the option to participate in hands-on sessions. Working in smaller groups they discussed, designed, conducted and presented small-scale research projects using the online eye tracking tool by RealEye.

More information about the summer school can be accessed on the dedicated website.


HAITrans to chair panel at the 10th European Society for Translation Studies Congress

29.06.2021

The European Society for Translation Studies (EST) scientific committee has selected the panel proposal submitted by HAITrans members Dragoș Ciobanu and Alina Secară, and colleague Julián Zapata from the University of Ottawa, to be part of the 10th EST Congress: Advancing Translation Studies. The 10th EST Congress is organised by UiO & Oslo Met, Oslo, Norway and it will take place from 22nd to 24th of June, 2022. Panel 27 “Past, present and future of speech technologies in translation — life beyond the keyboard” invites contributions from industry practitioners and academics that discuss observed advantages and disadvantages of integrating speech technologies into translation, PEMT, audiovisual translation, revision, or review processes; creative ways of achieving such integrations; novel training approaches created for such new integrations; as well as future directions of research, development and training.

More information about the EST Congress can be accessed on the dedicated page.


HAITrans at EDUinLOC - Europe Edition: A conversation between industry and academia

24.06.2021

HAITrans Senior Scientist Alina Secară and Esther Curiel, Localization Operations Manager at Indeed, were invited to share and exchange experiences about education and the localization industry at EDUinLOC – Europe Edition: A Conversation between Industry and Academics. EDUinLOC (a LocWorld initiative) is a global platform to enable communication between academics and the localization industry.

You can find more information on the dedicated event page.


HAITrans present at the #TranslatingEurope Workshop

04.06.2021

HAITrans Senior Scientist Alina Secară shared her experience on working with cloud-based subtitling systems with colleagues, students and representatives of the EU Directorate-General for Translation in her presentation Technological Developments: How to go about them in Subtitling Training at the Translating Europe Workshop - From Translation to Accessibility: EMT Train the Trainer Summer School. Technologies and Practices in AVT and Media Accessibility Training. More information is available here.

The event was organised by the Constantine the Philosopher University in Nitra, Slovakia - Kaunas University of Technology, Lithuania - Instituto Superior de Contabilidade e Administração do Porto, Portugal - Università degli studi Internazionali di Roma, Italy - University of Vienna, Austria and University of Antwerp, Belgium together with the Directorate-General for Translation of the European Commission.

A recording of the session is available here.


Workshop on Subtitling at the Nitra Summer School

03.06.2021

HAITrans Senior Scientist Alina Secară was invited to lead a workshop on subtitling at the Audiovisual Translation: Translation, Technology and Inclusion Summer School which took place online between 1-3 June, 2021 and was organised by the Constantine the Philosopher University in Nitra, Slovakia. Participants were able to use a cloud-based subtitling system and apply specific techniques to produce subtitles as part of a typical workflow.

More information about the event can be found here.


RAUN class of 2021

03.05.2021

UniVie HAITrans research group member Claudia Wiesinger has been selected as a participant for this year’s Regional Academy on the United Nations (RAUN) under the theme “COVID-19 recovery: towards more resilient and inclusive solutions”. In the upcoming months, she will get the opportunity to attend workshops and lectures, as well as co-author a policy paper on one of the many challenges caused by the pandemic. The Academy and the research project will be guided by practitioners, as well as experts from the UN and other partner organisations.

More information on the RAUN 2021 is available here.


HAITrans present at the #AFFUMT 2021 Conference - 8-9 April, 2021

09.04.2021

Dr. Alina Secară and Univ.-Prof. Dragoș Ciobanu shared their research on the ways in which speech technologies can improve the work of professional translators in their presentation at the #AFFUMT 2021 Conference: Translator Training: From the present to the future.

In their presentation "Benefits of speech technologies for translators and revisers", the two UniVie HAITrans research lab members shared some of the findings of their research project in which professional and trainee revisers who had access to automatically-generated recordings of source text segments identified and corrected more errors (especially Accuracy errors) left in the target text by previous translators. 

More information is available in this open-access publication which you are welcome to read at your leisure.


EMT Network Meeting - 24-25 March

25.03.2021

Univ.-Prof. Dragoș Ciobanu moderated the "New degrees preparing for new profiles" panel discussion which opened the second day of the EMT Network Meeting "EMT - breeding ground for digital linguists?". The panel shared good practices and insights into creating innovative training programmes at both BA and MA levels which respond to the latest industry needs, but also anticipate the types of skills which translators and intercultural communicators will need to succeed in the future.