A fresh look at remote simultaneous interpreting

An AIIC pre-candidate shares her thoughts and learnings from the AIIC Netherlands seminar on RSI

The Girl with a Pearl Earring

Photo credits: Johannes Vermeer / Wikimedia Commons

By Tatiana Kaplun

In early November 2018 I found myself, for the first time in my life, stepping off the plane in Amsterdam, and going, also for the first time in my life, to the Hague. It was also the first time in my life that I found myself so utterly and completely overwhelmed by a sheer wealth of information on remote simultaneous interpreting. And such balanced information too.

The AIIC Netherlands seminar on RSI seemed the perfect excuse to tick another country off the list, and it did not disappoint. So now, at long last, a brief recap (i.e. long read) of what I learnt and discovered in the Hague – apart from the Girl with the Pearl Earring portrait, of course. That I posted all over my social media right away, without waiting almost a fortnight.

But here's one more time, just because she's so beautiful to look at.

So, back to business. How do you normally start a proper story?

Let's go with once upon a time, like they did in the good old days…

Once upon a time simultaneous interpreting (as is, without any “remote” prefixes) was considered a disruptive technology, a threat to the profession, then it all calmed down a bit, people went with the flow… and now, as Sylvie Nossereau put it, “the profession is at the crossroads again”. One might even say that this comes as no surprise, but rather a logical progression of the gradual distancing of the profession. Just think about it, we went from consecutive side by side to the participants, to simultaneous behind glass, and now more and more often we hear about this new option, “distance interpreting”.

There’s nothing new about remote interpreting

First of all, it’s important to understand (and I liked that that was one of the first things to be said at the seminar) that there is nothing new about distance or remote interpreting, as it has been happening all over the world (and on a rather large scale) on a daily basis for quite a long time now, mainly in the form of consecutive interpreting solutions in medical and legal fields. Some countries (Germany and the US being among the more notable examples) were among the first to try it out in the simultaneous mode.

The elephant in the room, i.e. the problem of defining remote (simultaneous) interpreting, was also dealt with early on.

The main problem here is that, more often than not, remote simultaneous interpreting is used as an umbrella term of sorts, a generic concept used to “describe all kinds of situations where one or more of the participants in a communicative event supported by interpreters are not at the same location as one or more of the other interpreters”. We all think we are talking about the same thing, but we all have somewhat different pictures in mind.

Configurations can vary greatly, and each configuration requires different solutions. "Things" that can be different from one configuration to another may include the interpreting mode (sim v. consec), interpreting strategies, cognitive processes and technology involved in the process.

Who are you calling ‘remote’?

Remote interpreting (RI) is just one kind of distance interpreting (DI), characterized by the interpreter having no direct view of the participants in the communicative event.

But then what do we really mean when we use the term “remote”, the speaker or the interpreter? Or both? According to the classification proposed by Klaus Ziegler, if it’s the speaker that’s “remote”, then we are probably dealing with teleconference interpreting (audio or video). On the other hand, if it’s the interpreter, then that’s remote interpreting.

There’s even – so it transpires – an official ISO definition of distance interpreting (with “remote interpreting” being mentioned as an “admitted term”) as the “interpreting of a speaker in a different location from that of the interpreter, enabled by information and communications technology”... which brings us to what most of the colleagues came to the Hague to hear and talk about: this technology, the solutions and configurations available, and whether or not time has come for us to start panicking (or, at the very least, become somewhat concerned).

And if you thought “the cloud” part of the process was easy and straightforward – since that’s what remote interpretation goes through, for want of a better phrase, – think again. Even there, configurations might differ.

Here’s a list of the main technologies and options for sound and image transmission:

  • Only audio
  • Webconferencing solutions
  • Videoconferencing solutions
  • “Broadcast” point-to-point solutions
  • Hybrid solutions, combining the traits and characteristics of all of the above –more often than not the case with current RSI solutions .

Sim v. consec, hub v. home

And from this (highly) technical point of view, the fundamental differences between sim & consec become ever more evident, as signal processing, cognition, and necessary equipment suddenly take centre-stage.

The big debate between two possible scenarios – hub v. home –leaves out the fact that they are both just one part of the picture. In order for the system to work, “every point connected to the communicative event needs to be addressed, monitored and controlled”.

Which is why one of the speakers, Rachel Breviere, who presented her own hub model (with 11 booths, piloted as a new solution in Paris last year), put such a great emphasis on the conclusion that even if the interpreters could stay remote, the technician still had to go, just to make sure things didn't go wrong on-site.

Also, in order for any such a hub to work, one needs not only a suitable solution that keeps everyone (relatively) happy, but also colleagues who are well informed and ready to take the plunge,  local technical support, and a good and stable cash flow. Which is why, perhaps, it is a solution only suited to certain markets.

And the existing practices also have a role to play. Which explains why there is such a difference between the perspectives in Europe  and the US – where remote interpreting has long been a fact of life. It was interesting to  hear first-hand experiences  from colleagues working on the American market. Some of the ideas seemed too… crazy, for want of a better word, others just too complicated… others still simply uncomfortable, but they were interesting to see  nonetheless.

Losing those obtrusive buttons

The general idea is to replace the traditional “hardware” consoles, and their “obtrusive buttons” (yes, that’s probably my favourite phrase of the seminar), with a laptop, and replace the infrared channels with BYOD (Bring Your Own Device) solutions. This way the audience would be allowed to use their own devices instead of those traditionally handed out at the entrance to conference rooms, then collected at the end, some inevitably lost or taken as souvenirs.

All of which is great, and sustainable, and modern, and all those other wonderful  words and things, but the basic technical requirements are not really as basic as they might seem.

The basic frequency requirements, for instance, include the need for the microphone and headphones both for speakers and for interpreters to correctly reproduce audio frequencies between 125 Hz and 15 000 Hz. However situations where the mic frequency response is not compliant are quite common, with people arguing that it should still be enough and perfectly all right to "hear" the speaker.

It might well be enough for listening, but when you’re trying to do that weird thing of listening, interpreting, and monitoring what you’re saying, all at the same time, then even a slight straying out of that range into less desired frequencies may make your job harder, your stress level higher, and have an impact on your overall performance as well as the audience’s experience. Acoustic feedback and acoustic shock protection are also important, as injuries suffered due to acoustic shock do already happen.

These requirements also apply to handheld, lapel, head-worn and any other type of microphone, as well as to sound input from external sources like laptops and videoconferencing systems.

Another important basic requirement is lip sync – “sound and image from a distant site shall be synchronized” – and I was very interested to learn the stats for that:

  • Sound behind image no more than 45 ms
  • Sound before image no more than 125 ms

Latency also matters:

  • Image and sound must arrive at the interpreter’s screens and headphones within 500 ms after being produces at the source.

And, even though this might sound as an obvious point, but as far as image quality is concerned, there should be no visible artefacts (e.g. blurring or freezing), and, according to ISO 20108, non-compliance with this particular requirement may mean that “simultaneous interpreting might have to be suspended”.

Good quality sound and image don’t come cheap – and don’t come light, with the average requirement for Internet connection / bandwidth being estimated at 4 Mbps for every HD video feed, audio included. So if you want three or four screens, for instance, do your math and see if your Internet provider will be happy to oblige. Or not.

Standards matter

Conveying these standards to clients is incredibly hard. In general the market doesn’t care about standards. Which is precisely why we must.

Developing new standards for remote simultaneous interpreting will be crucial. But the focus should not be exclusively on the technical side of this challenge. Working conditions, health breaks, team composition also need to be agreed on.

A short list of some relevant standards:

  • Simultaneous interpreting – Permanent booths – Requirements:  ISO 2603 (revised in 2016)
  • Simultaneous interpreting – Mobile booths – Requirements: ISO 4043 (revised in 2016)
  • Simultaneous interpreting – Equipment – Requirements: ISO 20109 (new 2016)
  • Simultaneous interpreting – Quality and transmission of sound and image input – Requirements (including distance interpreting): ISO 20108 (new 2017)
  • Conference systems – Requirements: ISO 22259 (new 2018, publication pending)

Speaking of technical requirements, if you’re going to work from home (or any remote location), make sure you apply the same criteria to your work space as you would to your regular booth – a comfortable chair, a functioning light, the necessary number of screens, a proper desk, a solid console, a good headset, and water. Most importantly, one should use (and that’s something I should have probably realized sooner) a dedicated laptop. Why? Because the moment you use your laptop in any RSI configuration, you’re instantly giving access to the tech team to all of your confidential data.

Who’s to blame?

Then there is also the age-old question of liability (i.e. who’s to blame if it all goes wrong?) and control (i.e. who controls the environment?) to take into consideration.

Liability being a legal concept (with financial responsibilities attached), the question of who assumes liability for damages is important.

Think of it for a minute. Say, for example, your connection is cut, or there are works on your street, and someone digs up something they shouldn’t have… or something as simple (and unforeseen) as a power cut? Thanks to the stats brought to us by Gaspar Obregon, we now know that there are fewer car crashes in Europe than there are power cuts.

So who’s to blame then?

You do not want your contract to read anything remotely similar to this:

(…) the interpreter will be responsible for the stability of the internet connection at the venue he is working from, or for any malfunction of computer (…) being used.

That is why, if you do decide to get into the business of RSI, you should always have back-up, including a redundant internet connection.

One might argue that not all on-site interpreting jobs meet all those standards and requirements, but in the case of RSI they become of paramount importance, as the general conditions are already far from ideal, so to have them deteriorate even further would be downright cruel.

If you can’t beat ‘em, join ‘em…

As Rachel Breviere so nicely put it in her talk, “the best way to deal with transition is to be part of it, instead of just putting your head in the sand”. These changes seem to already be upon us (at least, in some regions and markets), so it’s important to be part of that process, to make sure it happens on our terms. If we are one day forced to work in the world of predominantly RSI assignments, we should do so the way we like and know, so that we feel can ensure we interpret well, without killing ourselves and our brains in the process.

But who are we kidding. We’d probably all hate to see that day arrive.

Or maybe that’s just a more skeptical approach. Just as it is true that most of us prefer working in our “real” booths and on-site, it is also true that RI could open new segments, and some meetings that could never have be held with traditional interpreting solutions might actually come to take place.

As was repeatedly stressed by Naomi Bowman, sometimes conventional approaches simply don’t work. As interpreters, we provide a service, and, whether we like it or not, we need to respond to the market needs.

How do you make this happen?

Limiting your liability

You start by mitigating risks, and thinking of yourself as a business, because that’s precisely what businesses do. They mitigate risks. And that’s what interpreters should do as well, for instance, by considering a shift from the traditional per-day approach to a project one, or finding a way to factor in prep time.

Speaking of risk mitigation, what really caught my attention was the mention of two types of insurance: “general liability” insurance and “errors & omissions” insurance. I should probably get out more, but both came as a surprise, as did the requirement by some companies to sign a waiver that one or both of these kinds of insurance are not required by law in your jurisdiction.

Remember the importance of contracts. The purpose of the contract is to limit your liability and risk, so it would seem only logical that you focus on the clauses that push your risk and liability to someone else. Which is why (again, I was surprised to learn) there is such a thing as a “best human effort” clause, which is basically there to remind the clients that what we are providing is not an edited translation, but, lest they forget, in-the-moment interpretation.

Everybody is afraid of commoditizing simultaneous interpreting, so we should focus on the human part – and commoditize the technical side instead.

There seems to be a growing recognition from the professional community of the importance of having independent and objective research on distance interpreting. There have been some studies, but we are still lacking a large-scale study with reliable data.

We have to be educated, to explore, to understand what is happening on the market and what all these companies providing a variety of functional or not-so-functional remote interpreting solutions are doing, and, if we think they are doing something wrong – let them know how to make it right, and give them feedback.

As they kept saying at the seminar,

It’s your obligation to be compatible with the future.

And that's, probably, as good place as any to stop and bow out.

Thank you so much again to Sylvie Nossereau, Claire Parment, James Norman, Jonathan Faydi, Levan Totosashvili, Ricarda Gras, Zouchra Kasimova, and everyone else at AIIC Netherlands for organizing such an incredible event, and for inspiring me to write about it.

P.S. And if that wasn't enough, check out this saved Twitter stream (special thanks to Julia Poger for putting it together), you may be able to find more information there.

Tatiana Kaplun is an AIIC pre-candidate based in Strasbourg.

Recommended citation format:
Tatiana KAPLUN. "A fresh look at remote simultaneous interpreting". aiic.ca December 21, 2018. Accessed January 21, 2019. <http://aiic.ca/p/8725>.