Exploring Real-Time Interlingual Subtitling in Broadcasting: Insights from ZDF Digital
Julia Borschak from ZDF Digital discusses the role and workflow of real-time interlingual subtitling professionals in the broadcasting industry.
File
LTA ME Real-time intralingual subtitling professionals in the broadcasting industry.
Added on 10/01/2024
Speakers
add Add new speaker

Speaker 1: Hello everybody, my name is Julia Borschak from ZDF Digital and I'm very happy to be part of this symposium and to tell you more about real-time interlingual subtitling professionals in the broadcasting industry.

Speaker 2: First of all, a few words about me.

Speaker 1: I am working for ZDF Digital since 2013. I started as a student's help and now I'm an accessibility editor for subtitles, audio description, sign language and plain language. For about four years I worked as a real-time interlingual subtitler or like we name it as a re-speaker.

Speaker 2: Now a few words about ZDF Digital.

Speaker 1: ZDF Digital is a subsidiary of ZDF, translated Second German Television. It's like the name says, the second public national television program in Germany. We are not fee-financed and a private limited company. ZDF Digital was founded in the year 2000 and meanwhile we have about 250 employees. This is what we do, our service portfolio. We offer film and TV productions, motion design, virtual reality, social media is a big point and websites, apps, sounds and music. And of course we also offer accessibility services, especially for television, like timecode subtitles, means everything which is pre-produced and not live can be subtitled previously with a timecode. We are offering live subtitles and we are also translating live in German sign language. Furthermore, we offer audio description, in particular for fiction formats and documentations. And our nearest project is plain language for people with learning disabilities. In this lecture I will give you an overview of the production of live subtitles for the television and of the work of a real-time interlingual subtitler. We are producing live subtitles primarily for the news, because of the actuality it is not possible to prepare subs previously and most of the news are broadcasted live, so it's simply necessary to subtitle them live. Furthermore, we are subtitling different kinds of shows, magazines, sport events like winter sports competitions, Olympic games or soccer cups and so on. And of course special broadcasts, for example, when something unaccepted happens like, I don't know, like terror attacks or other incidents, because especially in those situations it is very important to reach everybody with information. In the past years we are producing continuously more. This year we are at about 200,000 live subtitled minutes.

Speaker 2: This is the workflow for television live subtitles.

Speaker 1: At first we are checking the schedule of a show and asking the editors about scripts or moderation texts. These scripts must be typeset, that means the text is adapted to the subtitled guidelines. We are using a software called FAB. When the videos are available before, the text can be adjusted to the video and to the audio and then the speech recognition software has to be trained with vocabulary. And finally, when the show starts, everything prepared is transmitted semi-live, that means every single subtitle is sent simultaneously to the audio and the missing content has to be produced with re-speaking. By the way, we are transmitting the subtitles via teletext, so you can choose from the television on a special site and there are the subtitles via teletext. It's important to understand the whole real-time subtitling workflow and there are two main techniques in real-time interlingual subtitling for the text entry. One is based on speech recognition systems, the re-speaking, like I talked about on the slide before, and the second is based on fast typing systems. I will only go more in detail and talk about the first technique, re-speaking. Yes, like you can see it on the slide, the re-speaker listens to a speaker and simultaneously repeats, reformulates and shortens, if necessary, what they say to a microphone. This microphone is connected to a speech-to-text software, which automatically recognizes the re-speaker's voice and turns it into written words. By the way, we work with a software called Dragon. The reason why we are working with re-speakers and not with fast writers is that for us the training is easier. We choose good and fast subtitlers with enough experience and they get to know the software and practice listening, repeating and correcting at the same time. When you want to become a re-speaker, you have to have the following characteristics. Ability to deal with stress and personal resilience is very important as a re-speaker, because there is constantly a big time pressure, for sure, it's always life, and furthermore you have to handle difficult speech styles, I don't know, like a strong accent or a complicated content, that's not easy. And it's also important to talk about the working hours, for example, we at ZDF Digital are producing subtitles for early morning news show and this re-speaker shift begins at four o'clock in the morning. This could be very challenging and exhausting, but you get used to it. And concerning the requirements above all, you have to have a big, big, big multitasking capability because you have to perform several cognitive processes simultaneously while maintaining concentration and accuracy. Yes, and to give you a better glimpse, I have got a video for you, you can see me in action and maybe also interesting for everybody who has no experience in re-speaking, as you can see it in the video, that also punctuation has to be submitted to the software.

Speaker 2: So enjoy watching it, it's very short.

Speaker 3: What I wanted to be when I was little? Definitely not what I'm doing right now. Sure, dream jobs like vet or astronaut, but honestly, you just want a job that's fun. With nice colleagues and definitely not a boring office job. I wanted a job that challenged me.

Speaker 4: And yes, it might not change the world, but I bring the world a little closer to others.

Speaker 2: Yeah, here I am again.

Speaker 1: Yes, this was a short insight view how we work and a few more facts. At the moment, this work at our company are doing four permanently re-speakers, permanently employees and about 15 temporary employees who are only subtitling. Every shift is assigned to permanent teams and they consist of one re-speaker and one or two subtitlers or for talk shows, also of two re-speakers. Yes, we are working seven days a week and the inquiries increase. That's because the importance of accessibility is increasing. Subtitles enable hearing impaired and deaf people to access media. And they also support all those learning a language, in our case, learning German. And subtitles are always useful wherever it is better to watch without sound. For example, in public transport situations. That's why we are live on air every day and happy to be involved with European partners in the LTA project to train professionals.

Speaker 2: Yes, that's all. I'd like to thank you.

Speaker 1: Thank you very much for your attention. And when you've got questions, please let me know. And yes, I wish you a nice day.

Speaker 2: Thank you.

Speaker 5: LTA, Live Text Access Universitat Autonoma de Barcelona SDI, Internationale Hoogschule Scuola Superiore per Mediatori Linguistici ZDF, Digital European Federation of Hard of Hearing People, EFHO Velotype SubT Access European Certification and Qualification Association, ECQA Co-funded by the Erasmus Plus Programme of the European Union Erasmus Plus Project 2018-1-DE01-KA203-004218 The information and views set on this presentation are those of the authors and do not necessarily reflect the official opinion of the European Union. Neither the European Union institutions and bodies, nor any person acting on their behalf, may be held responsible for the use which may be made of the information contained here. Thank you.

ai AI Insights
Summary

Generate a brief summary highlighting the main points of the transcript.

Generate
Title

Generate a concise and relevant title for the transcript based on the main themes and content discussed.

Generate
Keywords

Identify and highlight the key words or phrases most relevant to the content of the transcript.

Generate
Enter your query
Sentiments

Analyze the emotional tone of the transcript to determine whether the sentiment is positive, negative, or neutral.

Generate
Quizzes

Create interactive quizzes based on the content of the transcript to test comprehension or engage users.

Generate
{{ secondsToHumanTime(time) }}
Back
Forward
{{ Math.round(speed * 100) / 100 }}x
{{ secondsToHumanTime(duration) }}
close
New speaker
Add speaker
close
Edit speaker
Save changes
close
Share Transcript