Translators talk Trados

Meet the experts who want to share their personal insights on using SDL Trados Studio with you.

If you use SDL Trados Studio or you would like to know what all the fuss is about, a group of translators have got together to produce a library of videos and blogs containing tutorials and useful hints and tips just for you.

See what they have to say... 

At the last ATA conference I had a good conversation with Henry Liu, the current and very active president of FIT -- the "grouping of associations of translators, interpreters and terminologists" -- about what could be done to help translators have a clearer voice when it comes to how their technology is going to develop. I've been thinking about that conversation ever since.

Some of you will remember that at the beginning of 2014 and 2015, I did webinars with eCPD to determine what's missing in translation technology. I wrote about it in a number of publications, but you can see most of the results in this (rather unwieldy) Excel spreadsheet (2014) and from the follow-up event in this fairly comprehensive document (2015). Those webinars (and the developers) were really quite helpful, but more can still be done to express what we want and to urge developers to consider and possibly implement our visions.

There are many online discussion groups and fora focused on a certain brand of technology where users express their ideas and wishes -- and where developers also respond. But Henry and I felt we still needed a way to articulate and develop requests and then send those to all current (and future!) developers. I think the best way to achieve this is by creating an open space like a Wiki or something similar that would allow for exactly that.

Now you're thinking I'm going to give you a link to that Wiki so we can start to collect our ideas?

No. I really can't take that on, but I would love for one of you (yes, YOU!) to do it. It would simply require a little bit of thought about the structure of the Wiki (or whatever it might be), setting up some rules of engagement, and a little bit of oversight to see that those rules are being followed. What I (and Henry) can offer is to promote it -- both to developers and practitioners alike.

Do we have a deal? Let me know if you feel that this is something you can do. Your colleagues would be very grateful.

MT has let you down. Lilt can help.

Adaptive, interactive MT that will change the way your team translates. 

Try out Lilt Business free for 30 days. 

Imagine all the world's best dictionaries at your fingertips!

For a fixed monthly fee, on all your devices, integrated in your daily work applications!

WordFinder Unlimited -- One service, all dictionaries

Try free for 30 days, check it out at www.wordfinder.com. 

At a conference last year I spent some time talking with Oleksandr Bondarenko, CEO of (yes!) Ukrainian LSP Translatel, about his CAT for Grad project. It's a project primarily aimed at (so far: Ukrainian) college translation programs (with pre-negotiated free access to some translation environment tools, etc.), but it also has some interesting content for the already-established translator through videos on a number of tools. The majority of these videos is in English; for the others, it might not hurt to brush up on your Russian and Ukrainian language skills.

Another highly interesting (and still developing) resource is the SCATE (Smart Computer-Aided Translation Environment) project's website. This project started in March 2014 and is slated to run for four years. Its goal is to "investigate several aspects of translation technology and its usage in the translator's workflow." These aspects include "underexploited translation resources," "poor integration of speech recognition," and "overloaded interfaces" -- all items that most of us would recognize as needing some work. The project is headquartered at the University of Leuven, and at a recent conference one of the project owners, Iulianna van der Lek-Ciudin, gave an interesting paper on term extraction -- or the lack thereof  -- in the typical translator's workflow and the underlying reasons. I would encourage you to read through her paper as well as the rest of the site. They are presently working on a more comprehensive interim report, and I will let you know once that is finished.

Protemos Translation Business Management System

Clients and vendors, projects management, invoices and payments, business reports...

Free for freelancers, 3 month trial for agencies, try it now!

First things first, I want to wish a happy and prosperous 2017 to all my readers. The Tech-Savvy Interpreter is now entering its third calendar year of publication. I couldn't do it without you. So, thanks for subscribing and reading each month. The demand for information about interpreting technology continues to grow at a brisk pace. To meet that demand, this year I am launching a monthly video to accompany this column. You can get a true "first look" at this month's tech here. Please subscribe the channel and spread the word.

I've observed the interpreting tech space for years. I felt like a had a good handle on the technologies out there seeking to innovate in and even disrupt the interpreting market, particularly when it came to interpreting delivery platforms (IDPs). Well, in November of last year I learned about yet another IDP called Headvox. The brainchild of language services veteran Frederic Eppendahl, Headvox was designed, built and launched in France. Frederic began to design the platform in 2005, initially using Flash and JavaScript before eventually moving to WebRTC a couple of years ago. FYI, while much of the interface and website is available in English, there are still parts that are only in French.

What is Headvox?

The best way to describe the Headvox platform is "Skype with live simultaneous interpretation," at least if you were to use it for a web conference like I did for this month's video. Of course, there is more to it than that, but at least the comparison gives you a rough idea. At first glance, it is a web conferencing platform that allows people who speak different languages to meet online and communicate through simultaneous interpreters who also connect to the meeting online. But as I have described before in other columns, there are many different use cases for remote interpreting. Each one is a little different and may have radically different technical requirements.

Headvox is a slick platform that has also been designed to provide remote simultaneous interpretation for multilingual face-to-face meetings (i.e. interpreters are not on site but the delegates are). They have also built an iOS app that functions as a delegate receiver. Android devices can simply login to Headvox through the on-board web browser to connect to an interpreted audio stream. This is possible because the platform is built on the browser-based technology WebRTC. This is yet another product that converts tablets and smartphones into interpreting receivers. FM and infrared receivers aren't dead yet, but we can expect to see more and more conferences and meetings making use of delegates' own devices to listen to interpretation.

User and Interpreter Interface

Of all the multilingual web conferencing interfaces I have seen so far, Headvox is the simplest to use. Its design is clean and largely intuitive. One of my initial criticisms of the platform was that it did not have a way to indicate when you were listening to the floor or original audio. It was confusing and I mentioned it to Headvox. Well, shortly after completing the video, Derick let me know that they have added a "floor button" to the interface. With a virtual interface, adjustments and new features can be introduced much quicker than on a physical interpreter console. Additionally, the iPhone and iPad apps are very simple to use and were not confusing at all. However, if using a smartphone or tablet app, just like a traditional receiver, you can only listen to the audio. You cannot participate in an actual virtual meeting.

Logging in to Headvox is simple. Go to the webpage, enter the specific access code and you are in. Selecting your webcam, speakers and microphone can be a bit tricky as the dropdown selection menus don't always prepopulate with the devices available on your computer the first time you log in. I had to guess and then exit and retry. The second time, it recognized my camera and USB headset, which appeared in the dropdown menus. Others who tested the platform with me noted the same problem.

The buttons for muting the microphone and turning off the webcam are conveniently located. The interpreter can see the speaker as well as other participants. When the screen share feature is engaged, the interpreter can see the shared content as well. To share your screen, you do have to install an extension in Chrome, but other than that, all features are immediately available in the browser.

Under the Hood

I tested Headvox on Chrome -- the browser that has most enthusiastically embraced WebRTC and the one recommended by Headvox. Video and audio quality are dependent on the quality of the webcam and microphone being used by the participants. That said, the video quality appears to max out at 720p/30fps.

There is no volume control within the Headvox environment. The interpreter must use the volume control on the computer. This is an example of the tradeoff that comes when trying to keep the user interface as simple as possible. Adding more power and more features creates complexity. It is always challenging to strike the right balance.

On at least two occasions when testing the platform, I had a problem with the microphone not unmuting and my video freezing up. Both problems were resolved by logging out and logging back in again.

What Does the Future Hold?

Headvox is still in its infancy. It lacks the polish of web conferencing platforms that have been in the hands of hundreds of thousands of users for a few years and have benefitted from successive waves of user feedback and complaints. The version of Headvox I tested only allows for bilingual meetings. However, Frederic contacted me just before publication and told me that the next version of the platform allows interpretation to and from up to five languages with automatic relay into two. They have also incorporated a private chat feature, so the interpreters can have a backchannel for communication while working remotely. I'll plan on taking a fresh look once the new version is released and will report back in a few months.

Do you have a question about a specific technology? Or would you like to learn more about a specific interpreting platform, interpreter console or supporting technology? Send us an email at inquiry@interpretamerica.com.

Talk Business Anywhere with Cadence

With over $1.5 million in payouts in two short years, Cadence is the leading platform for connecting freelance professionals to business leaders abroad. Complete your profile and if you don't receive an invite to a job in 6 months, we'll give you $50. 

Claim your profile today at talkbusinessanywhere.com.

I have talked about the Iceni Infix PDF Editor (edition 264 of the Tool Box Journal) and its delightfully creative way to convert PDFs into a translatable format and back-convert the translation into a PDF. There are two relevant and interesting new features in this rapidly evolving tool.

One is that the already-promised plugin for Trados Studio has materialized. This allows you to upload your PDF to Iceni's servers for conversion into XLIFF and subsequent translation within Trados Studio right from within Trados Studio.

Also, the optical character recognition (OCR) process has been greatly enhanced. Until now it had been sort of an afterthought. It was available as an option that you could choose if the text-based conversion did not display adequate results (as it could not have with image-based PDFs). Plus the OCR engine they used produced inferior results.

Now, in the latest version, the Infix PDF Editor automatically recognizes whether you are converting an image-based PDF and, if so, sends it to the ABBYY conversion server, likely the best OCR-based conversion on the market (it's the one that Wordfast Anywhere and SmartCAT use as well -- see edition 253 for a comparison). The fee that Iceni has to pay to ABBYY will be passed on to you as the user.

A third new-ish feature: It's now also possible to convert to RTF rather than staying within the PDF format for formatting purposes (again, if you remember from the earlier article, the Infix PDF Editor offers PDF-internal editing). I had really encouraged the folks at Iceni to implement that because clients don't necessarily want PDF back even when they send PDFs to start with. But I'm afraid that the way this was implemented is not particularly helpful. Rather than trying to emulate the formatting of the original PDF, all formatting is stripped and only the text is transferred. This might be helpful for some cases, but probably not for what I had in mind.

Still, knocking two out of three new features out of the park is a pretty good batting average! 

50% discount on crossMarket premium subscriptions

Do not miss the opportunity to get the crossMarket premium membership for half the price until February 14, 2017. Of course, the bundle also includes the premium version of the Across Translator Edition. Using the full function scope you can make use of your sales potential as effectively as possible.

For prices and a list of all premium features, visit  www.crossmarket.net. 

Here's a recent translation environment tool development that's been very positive: user conferences specifically geared toward translators.

Two companies with quite a bit of experience in this field are Kilgray for memoQ and SDL for Trados Studio. SDL launches around 20 "roadshows" every year (it looks like there are none currently scheduled, but once scheduled they should be announced right here), and Kilgray has hosted the highly successful memoQfest in Budapest for a good number of years. (I remember telling the Kilgray leadership after the first successful event not to do them every year because it couldn't be sustained by the users -- boy, was I wrong). The next fest is scheduled for July.

Wordfast has also organized a couple of successful events in the last two years in Riga and Nice and are now scheduled to meet in Varna, Bulgaria in May.

Across has also had user conferences for quite some time but not for translators -- they had meetings with other user groups such as translation buyers and LSPs instead. Now for the first time they will host a user conference for translators. It won't surprise you to find out that this will take place in Germany, though where in Germany is still up in the air -- or maybe better, up to the users. There currently is a poll that gives you a choice between Stuttgart, Frankfurt, and Cologne. Ok, I admit I'm a little hurt that Hamburg is not on the list, but there is always a 2018, right?

So, what are user conferences good for (aside from ♫dancing♫)? To help the developers move in the directions you want them to move. And does it actually work? It absolutely can, as this recent blog post by Kilgray's new CEO shows. Or some of you were present when we voted on how to actually pronounce memoQ (like Suzie Q of course) -- and Kilgray has never strayed from that!