When I talked with Tommi Nieminen a few months ago about the OPUS-MT and OPUS-CAT products (you can find that article in issue 324 in the archives), he said -- much to my surprise -- that according to his testing, his products (the locally installed MT and the connector to CAT tools) are as good as or better than any other trainable machine translation engine.

Except one. ModernMT.

I have written about ModernMT twice now (editions 297 and 308 in the archives), and I want to continue with that because it's a unique product with a really interesting approach to adjustable machine translation, combined with an affordable price, an incredibly long list of supported languages (Albanian, Arabic, Bosnian, Bulgarian, Catalan, Chinese, Chinese Traditional, Croatian, Czech, Danish, Dutch, English, Estonian, Finnish, French, German, Greek, Hebrew, Hungarian, Icelandic, Indonesian, Italian, Japanese, Korean, Latvian, Lithuanian, Macedonian, Malay, Maltese, Norwegian Bokmal, Norwegian Nynorsk, Polish, Portuguese, Portuguese Brazil, Romanian, Russian, Serbian Cyrillic, Serbian Latin, Slovak, Slovenian, Spanish, Spanish Latin America, Swedish, Thai, Turkish, Ukrainian, Vietnamese), and a surprisingly small group of users.

I admit that I don't use it myself -- the one large and ongoing project I could use it for is handled in a tool that does not offer connectivity to the customizable portion of ModernMT 😪 -- but I have a strong sense that I might if that were different.

For us, the users, the process by which ModernMT works is intriguingly uncomplicated: Any translation request is first sent to a translation memory of your choosing, from which a now-customized request (essentially your source segment plus some similar target data from your TM) is sent to a generic machine translation (hosted in a number of data centers, including in Rome and San Jose), which then sends back machine translation suggestions that are likely adjusted to the style and terminology of your TM (provided that the data of the TM had similar enough information to the original request). This means that you don't have to worry about training the machine translation engine data per se; you just need to use a current translation memory to "massage" the MT suggestions.

After some unfortunate early pricing hiccups, access to ModernMT now costs $25 a month (with a free trial month) and connects directly with all the necessary features to Trados through an app in the app store, to memoQ through a plugin, and of course to MateCat ("of course" because both MateCat and ModernMT are owned by the translation and technology company Translated). Presently, the ModernMT team is working on a Chrome extension as well, which should allow for the use of any browser-based translation environment and ModernMT.

There are also a number of other tools that you can connect to the generic ModernMT machine translation engine, including CafeTran and the various Wordfast tools (for the latter, see this video), but without the customization feature, which sort of defeats the point, in my opinion.

I talked with Translated's CEO Marco Trombetti and Davide Caroselli, ModernMT's VP of Product, about where they stand in the market and what's new with the product. The two features that they pointed out as new are faster processing speed and a reconfiguration of the translation parameters to give greater emphasis on the automated "TM input."

Even more interesting were the usage numbers they shared. About 80-90% of their revenue comes from enterprise customers (their most well-known account, much used for marketing, is AirBnB), and there are about 1,000 paying freelance translators (the latter, according to Marco, with a month-to-month increase of 15%). That was actually more than I expected -- the last time I had spoken to Davide it had just been (and I quote) "a handful."

I'm curious to hear what the numbers are the next time I talk to him.

In our last column, we discussed how to find the ideal interpreting headphones. 

More experienced interpreters might also use their favorite headphones with a USB microphone and mixer for remote interpreting.

This column is excerpted and slightly modified from The Interpreter's Guide to Audio and Video, by Josh Goldsmith and Naomi Bowman.

Josh Goldsmith is a UN and EU accredited translator and interpreter working from Spanish, French, Italian, Portuguese and Catalan into English. A passionate educator, Josh splits his time between interpreting, researching and teaching through www.techforword.com, which empowers language professionals to make the most of technology.

Naomi Bowman has worked in the conference interpreting industry for more than 30 years, with a 25-year focus on remote interpreting. She is CEO of DS-Interpretation, Inc., a technology-focused conference interpreting company founded in 1972. Committed to raising the global standards of conference interpreting, she is a popular public speaker on RSI, helping interpreters transition to a virtual world, and the effects of technology on the language industry.

Find out how to prep terminology in a flash without pulling an all-nighter in Glossaries for Interpreters 2.0 LIVE from July 12 - August 3, 2021! 

The live edition of the course includes 30+ course videos, 20+ exercises, weekly office hours with Josh, a live group Q&A, a pop-up group to discuss what you're learning, a private podcast, and tons of new lessons. Grab your spot before the price increases on August 15!

Did you know that time perception varies from person to person? Only about one third of all people can accurately predict the passage of one minute with their eyes closed. "Time optimists" – the people in whose estimation time moves more slowly than it actually does – have unrealistic expectations about the number of tasks they can fit into a given time interval. For example, they may believe that they can finish a work assignment, answer e-mails, and update their social media accounts in a single hour. Overly optimistic time projections fail to take transition times, the mental commute between different tasks, into account. As a result, there may not be enough buffer time to complete projects, and deadlines may be more stressful than necessary.

Our brains are not capable of switching seamlessly from one task to another. Think of a situation when you are deeply immersed in a difficult translation, and you receive a text message from a family member about an upcoming event. You respond, but it can take up to 10 minutes to fully return to your previous focus. The same process repeats with every disruption, and before you know it, you've opened another tab in your browser and are reading about cocktail recipes for your family event.

Regardless of your time optimism, recognizing your brain's need for transition times benefits your daily work output. Expecting yourself to translate different projects back-to-back and underestimating how long the work will take sets the stage for quality problems and errors. If you struggle with deadlines or regularly end up working extra night shifts, it may be helpful to examine how long it truly takes you to recover from distractions, and how much transition time you need for switching to the next task. To be more productive, step away from your desk for a few minutes when you lose focus. Instead of mindless scrolling, you can stretch, water the plants, or start the dishwasher. I have found that leaving a "where was I" note on my keyboard before a break helps me jump back into my work without wasting time.

Freelancing can be stressful. Many translators struggle to find time for the long-term strategic objectives of their business. They know that building a sustainable business requires daily attention to marketing, networking, finances, and technology – but when is there ever enough time?

The key to addressing long-term tasks lies in distinguishing between urgent and important. In urgent mode, our time awareness goes out the window. We tend to tackle the first task that comes to our attention, typically an email or another type of message. We translate, check other messages, respond to social media posts, and absentmindedly attend to personal matters. Working in this mode is associated with unhealthy habits, such as skipping exercise, eating at your desk, and spending too many hours hunched over a keyboard. What's worse, urgent mode as the default keeps us stuck professionally, working for high-volume clients at low rates.

In contrast, a deliberately chosen long-term perspective allows for better work-life balance decisions and frees up much needed personal time. Paradoxically, marketing, creating presentations, or writing blog posts, the activities that give you access to different client groups and better pay, can feel confusing and strangely undefined -- the classic "slow day" with few tangible results. I will share a few tips for dealing with slow days in a future column.

To improve the balance between urgent and important work modes, it is helpful to gain greater clarity about your personal work circumstances, including your availability, project load, and household realities. Useful time awareness aspects include:

Dorothee Racette, CT has been a full-time freelance GER < > EN translator for over 25 years. She served as ATA President from 2011 to 2013. In 2014, she established her own coaching business, Take Back My Day, to help individuals and organizations solve problems related to workflow and time management. As a certified productivity coach (CPC), she now divides her time between translating and coaching. Her book Complete What You Started (2020) provides a blueprint for carrying big projects across the finish line. You can read her blog at takebackmyday.com/blog.

It is a major pet peeve of mine that many translation tools, and especially translation environment tools, are designed to create two different classes of users: technically advanced and the rest of us. Now, most features are typically available for everyone's skill level. But the features that deal with complex file formats, advanced search-and-replace features, or complex filtering are often reserved for users who are either not shy (or have the time!) to sit down for a couple of hours to create complex queries or who already have the required skills.

In a way this is not surprising. After all, it is more likely for a technically advanced user to reach out to the development team behind any tool and ask questions, give suggestions, or enter into any kind of dialogue. After all, they are the ones who have the confidence that they speak the required lingo. The result is that the development teams see these users as the "power users" and feels that they are owed particularly much in regard to new features or improvements of existing features.

I don't think that I have to explain much more, since every one of you knows exactly where you fit on either side of this divide, and you're either happy about the situation (if you are a technically advanced user) or frustrated (if you're the rest of us).

The latest version of memoQ (9.8) has a number of features that you can learn about right here, but there's one in particular that tries to address the problem mentioned above. I'm of course talking about memoQ's new Regex Assistant.

And what again is "regex"? First of all, it doesn't help that this seems to be a word no one can agree on how to pronounce. It's either /ˈɹɛˌɡɛks/ or /ˈɹɛˌdʒɛks/, and it stands for "regular expression." It's coded language to specify a search pattern -- kind of like the options you know from the Microsoft Word search feature, only that Microsoft (frustratingly) doesn't actually use true regex in that case. Many other tools do, though, including virtually all translation environment tools, which allow you to use these expressions for any action to search, search and replace, or highlight text in situations where you might need it.

MemoQ is no exception, and the new Regex Assistant can essentially be seen as a peace offering to the technically less advanced users to give them as many tools as possible to write the code that their geeky brothers and sisters have been able to use for a long time.

Interestingly, when I talked to Zsolt Varga and Tamás Rell (Product Owner and UX Designer, respectively), they mentioned twice that the Regex Assistant is not to be confused with power regex editing tools like Regex Hero or RegexBuddy. Those tools can naturally still be used by whoever wants to compose their regex queries within those environments (and then paste them into the memoQ regex environment).

The Regex Assistant offers a couple of assists in writing the regexes that you might need to find or filter your text, plus a couple of verification tools for the validity of your code.

The Regex Cheat Sheet is a list of common syntactical elements with their description in natural language (i.e., * stands for "any character" or \d for "any digit") alongside the ability to actually compose a regex with those selections. The Regex Library consists of more than 30 pre-composed regexes that the developers tried to carefully assemble by asking their users for the most often-used (or most-desired) regexes for their daily work. Selecting one of those from the drop-down list will enter it into the Find What field underneath. Here you can leave the query as is or alter it for your specific needs.

Underneath the Find What field you'll find the Testing Ground field, into which you can enter any text from any source or target field within a memoQ project with the appropriate buttons, or manually enter text and see whether your created regex works satisfactorily on that particular example. If you have either altered an existing regex or created one from scratch, you can then save it in your Regex Library so you can use it again next time.

MemoQ's team is quite aware that the number of precomposed regexes is not going to be sufficient for all users, but they hope there will be a lively exchange of these expressions in user fora and privately between users. I asked Zsolt and Tamás why they wouldn't actually open an official market or bartering place for these valuable strings. According to them, this was not what the users asked for in their fairly extensive user research, but I can imagine that we will see this become more interesting once the masses (i.e., us) get the hang of this so far virtually locked feature.

By the way, those conversations with users have been happening either via surveys or on the relatively new Idea Portal that memoQ has opened up. So far, the Idea Portal is open to users who have a number of years of support contracts under their belt (presently there are about 500 participants in the portal), but it should soon be open to any user with an existing support contract.

And if you're still not sure what this is all good for, you can find some more detailed information on where and when to use the regex feature right here and here.

If you think this is a positive review of the Regex Assistant feature, you're right. Even if you don't really have an interest in extending your search capabilities in translation environment tools, we should all be encouraged that this seems to be a recognition on the part of developers that we all need to be brought into the fray when it comes to the influence we choose to exert on our projects.

Plus, I almost forgot to mention that this version of the Regex Assistant is only the first iteration, with an even more feature-rich version coming with the memoQ 9.9 release in September.

Two things:

First of all, I'm thrilled (THRILLED) that my favorite event of the American Translators Association's (hybrid) conference in Minneapolis is back: The "Dictionary Exchange." We tried this for the first time a few years ago and were not completely sure whether it would work, but it was just fantastic. You can read some recollections from the first time right here. If you do, you'll find out that it really is not so much an "exchange" as it is a place to give away dictionaries that are gathering dust on your bookshelves and (independently of whether you brought dictionaries or not) where you can add as many dictionaries as you want to your own treasured collection. If you're not planning to come to Minneapolis, you'll be able to send me those dictionaries in Minneapolis and I will put them out for you. I'll let you know an address to send them to in due time.

Also, I would like to point you to a webinar that might be much needed by some of us (and maybe even many of us!). Masterword, an LSP in Houston, is organizing a mini-conference later this month (July 29th) with the title "Perils and Pearls of Living Online -- Is working on-line impacting your health? Can we change that?"

They've put together an interesting slate of medical and emotional health experts. The registration is only $10, and you can find more information right here.

Subscriber to the Premium version of the Tool Box Journal have access to an archive of Premium journals going back to 2007.

You can subscribe to to the Journal right here. Thanks!

If you would like to promote this electronic journal by placing a link on your website, I will in turn mention your website in a future edition of the Tool Box Journal. Just paste the code you find here into the HTML code of your webpage, and the little icon that is displayed on that page with a link to my website will be displayed.

If you are subscribed to this journal with more than one email address, it would be great if you could unsubscribe redundant addresses through the links Constant Contact offers below.

If you are interested in reprinting one of the articles in this journal for promotional purposes, please contact me for information about pricing.

© 2021 International Writers' Group