Panayota (Yota) Georgakopoulou is a leading audiovisual localisation expert, specialising in the application of language technologies in subtitling. She offers her services to high-profile organisations around the world as an independent consultant, advising on strategy, quality, tools, workflows, and language resource and data management.
Yota holds a PhD in translation and subtitling from the University of Surrey and has over 20 years of experience in senior management roles in the audiovisual localisation industry. She implemented the first university modules on audiovisual translation in the UK, published the first guidelines on subtitling templates, participated in multiple research projects on language technologies applied to audiovisual text, served as the Managing Director of the European Captioning Institute and most recently as Senior Director, Research & Int’l Development, at Deluxe Entertainment Services Group before setting up her own consultancy firm.
Yota has authored and co-authored over 30 publications and is a regular speaker at international conferences and events on translation. She serves on the steering committee of the biannual Languages and the Media international conference, and her research interests include audiovisual localisation, accessibility, crowdsourcing, post-editing, machine translation, speech recognition, quality standards, translation big data and the democratisation of translation.
At what stage in your professional journey, audiovisual translation and techniques became intriguing?
Growing up in a tiny town in a subtitling country like Greece, I always found films and subtitling intriguing and a window for me to the rest of the world before I got to travel abroad. I studied English literature and translation and when it came to deciding on my PhD topic I couldn’t think of a better way to spend the next few years of my life than on something I was so passionate about: subtitling. I worked in the industry simultaneously, setting up multilanguage subtitling workflows in the UK at the time of the DVD boom. I had the rare opportunity to bring my work to my research and vice versa, and I set to study subtitling norms in various European countries, especially on the reduction levels that take place from the audio to the written text in the subtitles. I was fascinated by subtitle template files, which are the cornerstone of multilanguage subtitling production, and tried to draft guidelines that would do justice to the different subtitling traditions in Europe and would offer a good basis for such multilanguage production. I have just published an abridged version of these guidelines I originally wrote as part of my PhD, as there still isn’t much published on the topic of templates. It’s been nearly two decades since their adoption and widespread use in the industry, so I wanted to provide an additional reference point to researchers working on subtitling guidelines today.
In this new era of technology, how do you perceive the role of translators when dealing with developments, for instance machine translation or software localisation?
The role of translators has always been that of building ‘bridges’ between cultures, providing connection between people. The use of technology does not change this, in fact it accentuates this aspect of translators’ work, making their added value more visible in the communication of the source text author’s intent, be that audio, video, game, graphics or text. Translators have always been among the first adopters of technology, originally computers and dial-up modems, word-processing software and tools like spellers, then translation memories and other CAT tools, and now language technologies such as automatic speech recognition and machine translation. The tools themselves inevitably have an effect on the translation workflows and the way in which translators work, but the end goal is still the same. The role of the technology tools is to help translators with repetitive tasks and in their research, so that they can truly focus on the creative and communicative aspect of their work, and incorporate in it the world context that machines have no understanding of.
“AV localisation industry is at inflection point.”
How important is the localisation process?
We live in a digital, globalised world, where the highest driver for growth in businesses is international expansion. Netflix today has short of 160 million subscribers, over 90 million of which are international ones – this is where the bulk of the company’s revenue comes from. Coursera subtitle their courses in English and translate them in more than 65 languages by crowdsourcing translations from their Global Translation Community. They say translation is the platform’s main selling point, as translated subtitles led to increased enrolments by 200-300% already in 2014. This is all thanks to localisation; localisation is the catalyst for growth.
What is your opinion about the use of a pivot language?
Pivot languages have traditionally been used to make translation workflows more efficient where talent is lacking in specific language pairs for the volume of the work that is available. In the audiovisual industry, subtitling template files have traditionally been in English, though typically for English source language content, which gets translated in multiple languages simultaneously. However English has also been used as the pivot language when translating from non-English source content to all other languages, for example, a Japanese film subtitled into Greek from an English subtitle template file. English has served the audiovisual localisation industry well as a pivot language over the years, as it has helped expand the pool of available translators, since more translators work with English, and has thus helped cater to increased volume demands ever since video became digital. It is now time, however, to think about its appropriateness again, as more non-English language content is made available internationally, frequently among language pairs that have no similarity, linguistically or culturally, to English-speaking countries, e.g. Hindi or Mandarin content made available in other Asian or African languages and countries.
What are the challenges that you face when dealing with problematic terms? How do you cope with them?
In the audiovisual industry we have very specific terminological needs. Unless the content being translated is scientific, which would involve the relevant terminology from the medical or other scientific domains, the issue of ‘terms’ in media content is mostly limited to what is frequently referred to as Key Names and Phrases (KNPs). In other words, a collection of proper names used in a film or a series, together with their official or client-approved translation in each language, as well as other commonly-used phrases that need to be translated consistently across episodes and series. Names often come with annotations, e.g. regarding the sex of each character, or the politeness level being used between characters. The latter is a must in concurrent translation workflows where multiple translators are involved. Names are particularly important in live scenarios as well, even if the work is done intralingually, for the deaf and hard-of-hearing audience. In live captioning news or sports for instance, a captioner would ensure that the dictionary s/he is using is populated with names of politicians, sports players, clubs, etc. depending on the content of the programme, so that they are output correctly by the software used during live caption/subtitle production.
“If translation is to be commoditised, it is so that it is democratised and can be accessed widely.”
What are the most crucial traits that a translator has to possess nowadays and what, in your opinion, lies in the future for AV translation?
It is hard to predict the future of AV localisation, as so much is changing so quickly these days. The industry is at an inflection point: language technologies, speech recognition and synthesis, and machine translation are implemented in subtitling in order to help language providers cope with the unprecedented volume of work that needs to be localised. Most likely, the market will be commoditised, and we will see a lot of bulk jobs being created in post-editing machine output. But we will also see a lot of other, more interesting jobs, in which translators interact with the technology, monitor the tools and their quality, and provide feedback to improve the tools’ output, so their job can be facilitated further. Quality is at the very heart of the discussion about the changes that are taking place in the industry today. I believe quality standards will be implemented, and there will be more transparent discussions about the quality levels on offer in the market. I also hope that audiovisual translators will be recognised and properly credited for their work, and that truly exceptional ones will reach the fame of authors and have followers looking out for their work. I also believe AV localisation will converge more with interpreting and with gaming and that we will see a lot of cross-pollination from these industries. I encourage translators to remain alert, responsive to the needs of their clients and of the market, curious and agile, to experiment and augment themselves with technology tools, to specialise so they can offer expert services, and reinvent themselves as cultural consultants and ambassadors.
How important do you consider terminological research for a translator, especially when working on audiovisual translation?
As I explained above, terminological search in AV localisation for media and entertainment content mainly relates to lists of names and phrases that are encountered in the video material, unless we are talking about documentaries. The template files translators typically work with also include annotations for text like lyrics, pop-culture references, etc. So on the one hand there is the issue of consistency and continuity, for instance by using the same translation for a character’s name, that also reflects the linguistic choice made in other media, e.g. in book translations if the film script came from a book, and doing so consistently across different episodes, seasons and translators. Subtitlers are required to research all proper names that appear in the material they are working with, including those of companies or products, so as to spell them correctly, and to properly identify and use the official translations for products such as books, films, songs, etc. that may be mentioned in a video. On the other hand, there is also the issue of rights, as in the case of song lyrics. If an official translation exists for a song in a given language, it cannot be used without rights clearance first. So yes, terminological search is very important in AVT as well, and a lot of effort is invested by providers of AV localisation services in building libraries of such terms.
Can you tell us more about your speech on “The many faces of translation” conference? Which were the key points?
The conference “The many faces of translation”, that was organised by the Directorate-General for Translation of the European Parliament on 18e and 19e November 2019, was an effort to discuss recent developments to translation in a global context, and focused on translation types that are among the more creative translation verticals, such as literary and audiovisual translation and the translation of videogames, showcasing how AI is implemented in these domains as well. In this backdrop, I presented on “The many faces of audiovisual translation” and discussed the different types of AV localisation services that are available in the market, highlighting their complexity. The sector has been witnessing unprecedented growth in terms of the volume of video material to be localised, mainly as a direct result of the growth in digital video production and consumption, which is increasing rapidly in non-entertainment sectors as well, such as for e-learning or product demoing purposes. Production models have been customised to suit the needs of the marketplace in the 21St century. Specifically, I discussed relevant trends and workflows utilised by AV localisation service providers, and focused on the pivotal role of language technologies. The latter have already been used to an extent in the AV localisation market, primarily for accessibility purposes, but now the entire sector is being disrupted by their application and new job profiles are making their appearance as a result.
“The tools themselves inevitably have an effect on the translation workflows and the way in which translators work, but the end goal is still the same.”
In your view, how do you see the evolution of translation and how can it be developed in public and private institutions?
We are already seeing translation in different verticals being commoditised, as is also the case with AV localisation. If translation is to be commoditised, it is so that it is democratised and can be accessed widely, as any human right should be. I believe translation needs to be at the core of any public or private institution, so that the right of communication is granted to everyone equally, irrespective of language or disability. Accessibility goes hand-in-hand and is included in this expanded notion of ‘translation’. I am a fervent believer in the value of multilingualism that the European Union embodies, as an expression of cultural identity, which ultimately helps to promote democracy, transparency and accountability, all of which are critical for both private and public institutions.
How do you view the developments for translators in labour market in the translation field? Any advice for the new generations of translators?
I mentioned before that the AV localisation industry is at inflection point. Not only are the jobs changing and new job profiles are emerging in the market as a result of the use of new technologies and tools, but this is all taking place at a very rapid pace too. Metadata annotator, for example, is a recent job profile, as is the role of live translator respeaker, et post-editing roles to do with either speech recognition or machine translation output. They are not very common job profiles yet, but they are due to grow quickly and provide bulk employment to language professionals. We will also see more language engineers in charge of customising the output of language technologies to client’s needs, and eventually content curators moi aussi. Project managers will need to specialise in workflows that use language technologies, while the use of synthetic speech will give birth to roles for script writers et sound engineers that are capable of tailoring their services to the needs of synthetic speech as well as natural voices. Another fascinating new role I came across is that of a director for accessibility and translation in the filmmaking process, referring to the person that coordinates the collaborative production of translated and accessible versions of a film. I am sure there will also be plenty of other new roles that we cannot yet imagine. My advice to translators is: to never stop learning; to be receptive to the needs of your clients and the changes that are taking place in the market; to experiment with technology; to be aware of your own competences and augment yourselves with the tools that will help you; to not only offer expert services to your clients but be able to communicate your value, as the key to unlock international expansion for their business. Be visible!
Which book, paper, project, etc. would you recommend to people, and especially translators, to read/ follow?
When it comes to AVT, the book on subtitling taught everywhere is the eponymous one by Jorge Díaz-Cintas and Aline Remael, Audiovisual Translation: Subtitling, which dates back to 2007 and an updated version of which is coming out in 2020. Routledge also publishes interesting handbook series, such as The Routledge Handbook of Audiovisual Translation (2018) and The Routledge Handbook of Translation and Technology (2019), both of which are recommended reading. Just by going through the bibliographies of these books, one is bound to come across other interesting material to study, depending on where their interests lie. There is also the European Association for Studies in Screen Translation (ESIST), a non-for-profit association set up by academics and professionals in 1995 to facilitate the exchange of information and resources, and to promote dialogue and professional standards in the teaching and practice of audiovisual translation. As of 2018, the association has also funded the publication of the Journal of Audiovisual Translation. The sector is vibrant with research and publications, especially since the turn of the century, and new research is typically presented at one of the two biannual conferences that take place on alternate years: Languages and the Media, et Media for All. On the industry front, the Media & Entertainment Services Alliance represents companies involved in the creation, production and distribution of media and entertainment content, and promotes communication and collaboration among its members through events and special interest groups, such as its Content Localisation Council. Information about the industry is also included in reports on the localisation market published by research firms such as Common Sense Advisory, Nimdzi, Slator and others, which can be helpful in terms of staying up-to-date with the latest developments in the market.
Is there a question which you would have like us to ask you?
I could talk for hours about audiovisual translation, but I think we’ve already covered many different aspects of it. I would like to thank you once more for the kind invite to do this interview. I hope it is of interest to the TermCord staff and I wish you all my best for the future in continuing the excellent work your unit is doing!
Antonia Pappa — stagiaire en communication à l’unité Terminologie
Né en Grèce en 1992. Elle est titulaire d’un baccalauréat en communication, culture et médias et a travaillé pendant trois ans pour un journal et des magazines alimentaires en Grèce. Antonia est maintenant titulaire d’une maîtrise en marketing international et communication et travaille à sa thèse sur la publicité sur les médias sociaux. Pendant son temps libre, elle aime voyager, faire du yoga et se promener avec son chien.