Voice or Machine Translation?

Post Survey Note: Thank you to all those who completed the survey. It’s no longer live, but you can see the final results in the article.

For the last couple of years I’ve been enjoying the TCLoc Masters degree at the University of Strasbourg.  It’s been a really interesting time for me helping to fill in a lot of gaps and widen my technical knowledge around localization, and introducing me to the world of Technical Communication in general.  This latter part was particularly interesting because half of our business at SDL relates to this; so having spent my time since 2006 working with our localization products it’s been an eye opener in many ways.  I have done this in my own time and not as part of my job, but TCLoc does look like a course that’s tailor made for SDL employees!

The course work is pretty much all over now and I’m working on my dissertation which is the reason for this blog article today.  I’ve chosen to question the effectiveness of voice recognition technology for dictation, compared to the use of machine translation, in particular neural machine translation.  I have many opinions around this of course, but what I’m really interested in is your opinion as a group of professionals who work with this technology every working day. To get this I’m hoping you’ll spend a little time to answer the questions in this article, and I’ll try to explain the reasons for each question as I go.  Each question is a simple checkbox, sometimes only one answer is allowed, sometimes multiple answers.  If you have something different to say I welcome your feedback in the comments.  There’s only seven questions, but thank you all in advance for sharing your views on this and helping me with this task.

An important point here is that this is not just for Trados Studio users!  I’m interested in the feedback of all professionals working in this field irrespective of the tools used.

The Survey

Do you or don’t you use dictation?

To begin with I’m interested in how many of you actually use speech recognition for your translation work.  Sometimes it sounds as though being able to dictate is something every translator wants to be able to do… but do they?

If you don’t use voice technology at all, then why is this?  Certainly in the last few years we’ve heard from a very ‘vocal’ part of our profession who use it regularly.

If you do use it then what is it that ensures you use this in preference to typing your translations?

What about Machine Translation?

I can still recall the times when the idea of using machine translation was considered laughable and very few professional translators would use it.  Many were curious, and of course worried about whether or not this technology would replace them, but it was mainly the butt of a good joke!  Today however things are different.  Most translation tools provide a variety of plugins for machine translation and it’s probably fair to say that most translators working with CAT tools today take advantage of it.  Or is it?

Some CAT tools provide the ability to work interactively with machine translation as well as supporting an ability to pre-translate and post-edit.  So if you do use machine translation, here’s a few questions… tick all that apply.

One of the benefits of neural machine translation is its fluency.  The quality of the sentences have become incredibly good, especially for some languages, and this is the main reason we are seeing so many new solutions appearing on the market and also why so many translators have adopted its use today.  But this is a double edged sword, because it’s much harder to spot the mistakes and neural machine translation will sometimes sacrifice accuracy for fluency.  Because of this the processes in place for quality assurance need to be really good, especially when post-editing.

One way to make this easier and still retain your productivity is to use speech synthesis.  What I mean by this is you can use voice technology in the form of text to speech to read the source, or the target, to you.  This way, for example, you could listen to the source and read the target simultaneously while you are post-editing.  Perhaps not a solution for everyone, but for professionals with the right skillsets it might be a useful productivity boost and assist with the quality control while post-editing.

So my questions here would be…

And my final question in this short survey is related to whether you, as a user of dictation solutions, think the ongoing improvements in machine translation will supersede the productivity gains you have seen through dictating all your work.  There could be a variety of answers around a question like this so just tick all that apply and of course feel free to add in any additional comments of your own.

Survey over!

That’s it… and if you made it this far and have completed the survey thank you very much!  Before you go, I added some sharing buttons at the end of this blog, so please feel free to share this through the social platforms you use and hopefully you can help me to improve the significance of the results with a larger data set to play with.

And of course if you have any opinions you’d like to share that won’t fit into anything I’ve said so far feel free to post your comments below… I’ll welcome the discussion.

AdaptiveMT… what’s the score?

AdaptiveMT was released with Studio 2017 introducing the ability for users to adapt the SDL Language Cloud machine translation with their own preferred style on the fly.  Potentially this is a really powerful feature since it means that over time you should be able to improve the results you see from your SDL Language Cloud machine translation and reduce the amount of post editing you have to do.  But in order to be able to release this potential you need to know a few things about getting started.  Once you get started you may also wonder what the analysis results are referring to when you see values appearing against the AdaptiveMT rows in your Studio analysis report.  So in this article I want to try and walk through the things you need to know from start to finish… quite a long article but I tried to cover the things I see people asking about so I hope it’s useful.

Continue reading

Spot the difference!

001I don’t know if you can recall these games from when you were a kid?  I used to spend hours trying to find all the differences between the image on the left and the one on the right.  I never once thought how that might become a useful skill in later life… although in some cases it’s a skill I’d rather not have to develop!

You may be wondering where I’m going with this so I’ll explain.  Last weekend the SFÖ held a conference in Umeå, Sweden… I wasn’t there, but I did get an email from one of my colleagues asking how you could see what changes had been made in your bilingual files as a result of post-editing Machine Translation.  The easy answer of course is to do the post-editing with your track changes switched on, then it’s easy to spot the difference.  That is useful, but it’s not going to help with measurement, or give you something useful to be able to discuss with your client.  It’s also not going to help if you didn’t work with tracked changes in the first place because you’d need some serious spot the difference skills to evaluate your work!

Continue reading

Qualitivity… measuring quality and productivity

01In the last year or so I’ve had the pleasure of watching Patrick Hartnett use the SDL Openexchange APIs and SDK to develop SDLXLIFF Compare, then Post-Edit Compare, the Studio Timetracker and a productivity tool that combined all of the first three into one and introduced a host of productivity metrics and a mechanism for scoring the quality of a translation using the Multidimensional Quality metrics (MQM) framework.  This last application was never released, not because it wasn’t good, but because it keeps on growing!

Then last month I got to attend the TAUS QE Summit in Dublin where we had an idea to present some of the work Patrick had done with his productivity plugin, get involved in the workshop style discussions, and also learn a little about the sort of things users wanted metrics for so we could improve the reporting available out of the box.  At the same time TAUS were working on an implementation around their Dynamic Quality Framework (DQF) and were going to share a little during the event about their new DQF dashboard that would also have an API for developers to connect.

Continue reading

The ins and outs of AutoSuggest

001The AutoSuggest feature in Studio has been around since the launch of Studio 2009 and based on the questions I see from time to time I think it’s a feature that could use a little explanation on what it’s all about.  In simple terms it’s a mechanism for prompting you as you type with suggested target text that is based on the source text of the document you are translating.  So sometimes it might be a translation of some or all of the text in the source segment, and sometimes it might be providing an easy way to replicate the source text into the target.  This is done by you entering a character via the keyboard and then Studio suggests suitable text that can be applied with a single keystroke.  In terms of productivity this is a great feature and given how many other translation tools have copied this in one form or another I think it’s clear it really works too!
AutoSuggest comes from a number of different sources, some out of the box with every version of the product, and some requiring a specific license.  The ability to create resources for AutoSuggest is also controlled by license for some things, but not for all.  When you purchase Studio, any version at all, you have the ability to use the AutoSuggest resources out of the box from three places: Continue reading

Language Cloud… word-counts… best practice?

001Best practice!  This is a phrase I’ve had a love/hate relationship with over the course of my entire career… or maybe it’s just a love to hate!  The phrase is something that should perhaps be called “Best Suggestions” and not “Best Practice” because all too often I think it’s used to describe the way someone wants you to work as opposed to anything that represents the views of a majority of users over a long period of time, or anything that takes account the way different people want to work.  In fact with new technology how can it be “Best Practice” when it hasn’t been around long enough in the first place?  I think for a clearly defined and well established process then “Best Practice” has it’s place… but otherwise it’s often the easy answer to a more complex problem, or just a problem that is considered too hard to address.
Continue reading

Solving the Post Edit puzzle

#03It would be very arrogant of me to suggest that I have the solution for measuring the effort that goes into post-editing translations, wherever they originated from, but in particular machine translation.  So let’s table that right away because there are many ways to measure, and pay for, post-editing work and I’m not going to suggest a single answer to suit everyone.

But I think I can safely say that finding a way to measure, and pay for post-editing translations in a consistent way that provided good visibility into how many changes had been made, and allowed you to build a cost model you could be happy with, is something many companies and translators are still investigating.

Continue reading

"Memory is the mother of all wisdom"

#01I believe this interesting quote can be found in “Prometheus Bound”, a play by a Greek dramatist called Aeschylus.  I haven’t read the play, but I like the quote, and it certainly lends itself to the importance of memory… even when we refer to a Translation Memory rather than your own built in capability.  It’s because your Translation Memory is such an important asset to you that you need to regularly maintain it, and also reuse it wherever possible to expand the benefits you get from it.
Continue reading

There's more than one way to skin a CAT

Updated: 14 January 2015
Today SDL is all about SDL Language Cloud and not BeGlobal, but I hope the article is still as relevant today.  There are more ways to look at how you use Machine Translation so if you’re interested take a look at these two more recent articles as well.
The ins and outs of AutoSuggest
Language Cloud… word-counts… best practice?
The title of this post could be quite tricky to translate in many languages because not everyone uses the expression in the same way, and certainly don’t use the same words.  I chose this especially because I thought I’d write a little about using Machine Translation in SDL Trados Studio.
I’m not going to talk about properly trained Machine Translation engines such as SDL BeGlobal, which can be configured and improved to provide remarkably good translations in a short period of time for very large numbers of words… so achieving economies of scale that would be unthinkable with human resources alone.  Instead, I’m going to talk about how a Translator can make use of the growing number of Machine Translation resources in a way that might make sense for them.
Continue reading