How can translators remain successful in an MT-driven world?

Igor Afanasayev 00:06
Our next presentation will be from Tess sweetie, who is a certified English to Swedish translator, translator and localizer. And she is also an author of few books that you can find on Amazon about marketing and linguistics. So, as I was looking through the description of the presentation from tests, there were a couple of like a specific phrase that caught my attention, which is human translators now, just think of it 15 years ago, or 10 years ago, if you would hear that or see that, in the context of some some paper, you would think that you are looking at some, I don't know, like, some sci fi novel, instead of like the actual description of the presentation. But here we are living in this future and trying to compete with machines and trying to find our place in this in this new world. So test today we'll be talking about exactly this, this trend and how to navigate the role of a linguist in the world where machine translation is becoming, you know, more and more important. So we are having tests here. Let's see if she has her microphone and video Connect.

Tess Whitty 01:24
Me. Yes,

Igor Afanasayev 01:26
yes, we do. We do. Will you see? Sorry? Will you see your as well? Hope okay, you will be you will wish you all the sharing your screen? Oh, yeah, we just see that. So please, please start with your presentation. I'm really excited about that.

Tess Whitty 01:45
Okay, and you can see me and hear me. Exactly. Good. Well, thank you, Igor, for having me. And thank you to everyone here. I know, I have a lot of familiar faces here that have told me they will be attending. And it was fascinating to hear about micro and macro trends from Renato, just before this. So thank you to all the organizers to it's a great event. And to now I'm going to talk about machine translation, because no one can deny that machine translation isn't here to stay, and that it's getting better and better every day. So the question is, will we need freelance translators or human translators eager in the future, and for what, that's what this talk, the focus of this topic is. And for those of you who don't know me, my name is Tess, sweetie, I am a human translator. I've been a translator for 17 years into Swedish. And I'm specializing in digital marketing. And I have a marketing background. And I am also a marketing trainer for other translators, and a podcaster, etc. So why am I giving this presentation? So as a trainer and a translator, I have been curious about machine translation ever since we started hearing about it. It's gone from mockery to feeling like a threat to adapting to it. And I feel obligated to stay up to date on it because I get a lot of questions from other translators about this topic too. And I also think that it's important for everyone here, translation buyers, and translate en and translation companies to hear this perspective of us humans, translators. So the presentation is based on my own experience, but also on the general opinions of my colleagues and from articles and research from industry associations and posts. So let's get started. What I wanted to cover in the short time we have together today is the current market situation for us OS translation, translators and translation companies. I want to talk a little bit about if MT could outperform humans, and when to use machine translation and when to use human translators, and the most important of all how we translators can remain successful. So let's first take a look at the current situation for translators and language service providers. I am sure you all are aware of these things, but there are some trends. I wanted to highlight and point out. So the demand for translation and services or translations is at a record high, but inflation adjusted earnings are declining. And I wrote still here because the pandemic has also changed a lot, but the demand is still there for all of us, so that is all good. And according to the Bureau of Labor Statistics, now, this is from 2019, or pre pandemic, the IP employment outlook for translators and interpreters, has been projected to grow by 29% through 2024. I don't think the pandemic affected this very much, but we will live to learn about it. This is also stated in an ATA article from 2019 called Is there a future in freelance translation? Let's talk about it. But our expenses go up just like in any other industry, but the rates have remained stagnant or decreased. The economists claim that translation is a growth industry. The problem is that the growth is in volume and not necessarily rates. So this means that cheap service is growing faster than quality service. Why is this? Now this is not the case for everybody. But this is a general trend. So the globalization and the technology like that Renato also talked about, and the growth of demand has also increased the supply. So an increasing number of people have started offering their services as translators. And like all professionals affected by the global competition these days, most translators in in in high income or high cost of living countries have a hard time competing with the same translators offering their services in lower income countries or a lower cost of living countries. The the trend is also or the fact is that we have a very unregulated profession, just like many other creative professions. There's no standard career paths stipulating the minimum level of training or experience or credentials required. And as long as the profession remains open to anyone and everyone competition will remain high, and keep can keep the prices low. But one of the most important factors of all this is that clients are either unaware not not all clients, but many clients, they are unaware that all translations and translators are not created equal. Or they are simply unconcerned about the quality of the service they receive. This is usually because they don't understand the importance of a good translation. So these clients will not be willing to pay a premium price for a service they don't consider to be premium. And I am sure that many language service providers are aware and struggling with this too. And lastly, then we have machine translation, which which will be the focus of today. So machine translation is what's driving, post editing or machine translation business more and driving the prices down. And more and more language service providers are relying on machine translation today. So the question is, can machine translation outperform humans? So I recently stumbled upon an interesting article or a paper called Achieving human level translation. And this was an article in a paper published on September 1 last year 2020, about kubot, a new transformer based deep learning system. And the authors were researchers from Google and Google Brain, Berlin and many others. And they said, achieving human level translations is so to 2018. In 2020, it has become outperforming human level translations. They also say and that's what I've quoted on the presentation. Highly qualified human translators with an infinite amount of time and resources will likely produce better translations than any empty system. But many clients cannot afford the cost of such translators and instead use services of professional translation agencies where the translators are under certain time pressure. Our results show that the quality of professional agency translations is not unreachable by machine translation, at least in certain aspects, domains and languages. So, this doesn't sound very good. And this is not what I am saying. This is what was said in this article. So the question is, are we going to be replaced by machines? And my short answer is, I don't think so. After following this for over 10 years, my current stance is this. There is a need for both machine translation and professional high quality human translators. So there is no way human translators could cover all the whole, all translations that are needed today. So if we need both, both machine translation and humans, when should we use machines and when should we use humans? This depends on of course, the complexity and the priority of the translation. So for example, if you're looking for to translate a lot of content in a really short time, then machine translation can outpace human translators and bring massive projects into more and more realistic timetable. But we don't know about the quality. Also, the simplicity, which comes to the simplicity, when translating repetitive information with clearly defined language terminal terminology, machine translation can handle the repetitive and simple texts. For example, there's no need for me anymore to translate the names of the countries or languages, and then that we come to the content priority. Low Priority content will be perfect for empty when you only need to understand it. So content that will be used internally might not require the same level of polish, other landing page for a new product, for example. So the focus of this presentation is professional translators and how we can remain successful. So like I said, previously, there is no way all the human translators in the world could handle the demand for translations these days. We do need machine translations. But we translators can no longer stick our head in the sand and pretend that machine translation won't affect us. Instead, we need to have an open and positive attitude and keep an unbiased perspective. To see how we can fit in and benefit from it. We need to keep up to date and we need to be interested in the developments. So there is room for both, like I said machine translation and professional human translators. But if we human translators, then we'll remain successful we have to focus on this segment here. The high quality translations are those that require the expert touched of a practice to translator with a full command have the context, human intent and target audience. This also means that if you are focusing on quantity as a translator, instead of quality, you cannot really compete with machines anymore. So to compete with with machines, here are some suggestions on what we translators should focus on. We have to be more of a specialist than a generalist and machine have a has a harder time competing with specialist knowledge. And clients. Also, our clients translate a client's language service providers and end clients are of course more likely to look for translators whose skills match perfectly with the job they have in hand. And instead of searching for a good all round translator and like we saw earlier in when to what humans have to focus on, we need to focus on more high priority and critical content. For example, some high at stake legal documentation or medical information will always require the expertise of a specialist human translator And also, we can focus on impactful published materials published that cannot, you cannot go in and edit. And that has a high impact on the bottom line for a company, for example, these are publications that need to be perfect. These are, for example, annual reports, pharmaceutical research journals, or any published material that can affect a company's image or that are trying to sell. And we also come to transcreation transcription is the art of creative translation. It's a service in itself because it requires deep consideration by a marketing minded translator. And the goal is to ensure that your message keeps the same impact and meaning across borders or that it's localized to the the target market. It also requires adaptation, perhaps of an entire tone, to ensure that the message and and goals are consistent. It's also not only length, language related. Audiences react differently within their cultures to certain things as numbers, structures, images, and even colors. So machine translation will not be able to combine culture, language and emotion into a creative and compelling text anytime soon. For this, you will really need to use human translators. But there's also a growing area of specialization to consider there, the developments in the technology and the globalization has meant that there's a huge rise in the demand for digital marketing of translation services, international companies realize these days that machine translation tools are simply not up to the challenge of optimizing a website and digital content for new markets. And they are looking for native translators with experience in digital marketing, so that machines cannot do keyword research or create compelling meta tags or localize the actual message on the same way on a website. So, if we want to compete with machines, we have to also move up in the value chain by focusing on high value areas which I spoke about. But also another idea is to offer value added consultancy services. So this can be becoming more like a brand or cultural ambassadors. So when companies try to personalize their offerings, they're trying to sell with emotions, offer and adapt to the tastes of the target countries, they will increasingly depend on local ambassadors or consultants who can advise them to the cultural specifics of a given target community and help promote products and services in this local language. So translators can therefore also be aware of the new opportunities that can open up due to this due to their knowledge of a specific language and the related culture and we can offer services to our clients to help build trust among the local customers and on the ground. And of course, we can't forget about post editing is become it's becoming a subcategory of translation, I think post but it is a viable service option for freelance translators. I have been asked to do it several times. But as an old timer, I have taken the stance to not do this but to focus on the other services instead. And I'm still kept very busy. And post editing make sense when the training data of the machine and the quality of the domain and is very well defined. But if it's not any productivity gain by using machine translation engines is negated by increased professional human editing time, if you need to post that if it needs to be really good quality. And this is also discussed asked in the ATA position paper on machine translation called a clear approach to a complex topic. We human translators should also remember that we can use machine translation to our advantage in certain cases, professional translators, we can use it as an additional tool or resource to improve crease translation speed and consistency. We can use it as a reference material or as a source of alternative. Try translation memory. So suggestions. And for example, if I have a long list of countries, I gave the same example earlier, there's no need for me to manually go in and translate all the names of the countries anymore. That is what a machine can help me with. And you and that's what I could use it for. So that's just an example. I'm not saying that I'm should use machine translation to produce. I can't use it to produce high quality translation, but I can use it as an additional tool in certain instances. So to summarize, all this, we human translators need to up our game, we need to focus on niche markets we use, we can use machine translation as a way to boost productivity in certain instances. And we can provide value added services. I recently also read an article from smart Ling that I think summarizes this presentation well, or my topic, human translators are the ones that will provide that extra level of creativity for the highest quality of work. Machine Translation is all about getting the job done with a margin for error. And this is a quoted by an article from an article by Matt Gray, the Content Marketing Manager at SmartLink. And here are some references used the ATA position paper on machine translation. And Taos article. What options do translators really have another ATA article, is there a future in freelance translation? Let's discuss, or let's talk about it, the SmartLink article, the state of machine translation, and the paper in Nature Communications, transforming machine translation, a deep learning system reaches new translation quality comparable to human professionals. So with that, I want to Oops, here are the references. And with that, I think you can get an access to the presentation with the links. If not, you can email me and I can send you the links. And here's my email and my website. So with that, I want to leave it over for any questions we might have. And I wanted to say, since I'm focusing on transcreation, as one of the solutions where we can up our game, I am giving away the Get Fit for the Future of transcreation by Nina Sackler have dar today.

Igor Afanasayev 23:28
A Awesome, thank you so much for your presentation, I was just going to announce that you will be getting the book that but you did that yourself. We already have a couple of questions. And let me first of all, first one was not really a question but a comment about you sharing your presentation. I believe you covered that as well. The best way to share your presentation would be going to our community where he can answer more questions from the audience, as they will be watching the recordings. And also they will be able to download your presentation there as well. Another question or comment was specifically about the availability of recordings those will be available to the to the people who were signed up for the conference. So if you will, of course it was a in is still going as a 12 hour marathon and not everyone can see all the presentations live so we will have all the recordings. So the question from Katherine Bosman is what kinds of value added services have you seen freelancers leaders offer successfully? Can you give some specific examples?

Tess Whitty 24:38
Okay. So one I can give an example from I gave the example of cultural consultations I have seen translators do you offer that as a value added service? Personally, I am studying right now. A lot. I'm partly studying copywriting so I can become a better trance crater, I am also studying international search engine optimization. So I can help with, with with that when in digital marketing when I localize or transcribed or localize a blog post or social media posts and I have also localized a lot of Pay Per Click ads. And then they need as a value added service I need to know or be good at Search Engine Optimization and find the right keywords for the local market. So those are some examples.

Igor Afanasayev 25:40
So much, there's no question from Martin chiastic. I'm not sure if I'm pronouncing his name correctly. What do you think about trainees or new translators, so they are usually the ones who provide who are accepting the jobs for low rates, what will happen to them in this new world where we have to compete with machines.

Tess Whitty 25:58
So first of all, if you are a new translator, I recommend getting as much experience as possible. In in a good way, so So volunteer translate a little bit every day, you don't necessarily have to lower your fees. If you start out with a low fee, you will be have a harder time increasing your fee, your rates from the beginning. And if you start out with really low low rates just to get get the job, you might not be taken as seriously that said, you do need to get your experience in. So as soon as you get a job, try to get as much feedback as possible. Translate, not for clients, but on your own and have a professional translator, review your translations just to get better and so that you know that you can provide high quality translations. And as soon as possible, try to focus on one area of specialization that you can get really good at. So I don't know if accepting really low rates is the way to go. Even for beginning translators, I understand that it's hard to get into the market, I would focus on getting as much training, continuing education and experience as possible, but not necessarily go for the low rates, but you need to up your game you need to get become good at translation, and be become an expert in your field in order to be able to compete with the machines. Hope that answers your question.

Igor Afanasayev 27:43
Thank you. I hope it does. The last question actually will be from me, I guess I guess they want that book badly. What? What do you think about people being shifted towards not providing translation directly but being like, empty translator empty training specialists, people who can improve the quality of machine translation over time by producing really quality signal or really quality translation, which hopefully will pay more for them?

Tess Whitty 28:13
So you mean that a translator would shift over to train the machine instead? Yeah. I guess that's another another job. But then you're not? Yeah, it's sort of the same as the post editing. It's sort of a subcategory. So the if you become a trainer of the machine translation, you you that's another subcategory. Do you mean, is it as a viable profession? Is it a way to go for us? Or

Igor Afanasayev 28:47
is it something that a linguist could could shift towards to like being a really professional linguist, knowing how to do how to provide translations, the best way to know in order to really good at good signal to improve the quality of machine translation over time, like, like cherry picking the things that are really needed to, to up the up the level of the overall translation, but like, exactly knowing where and what to do in order to improve the quality?

Tess Whitty 29:16
Yes, the short answer is yes, I think so. That's, that's another career path. That's another way a professional can become a machine translator trainer. i Yes, I guess so. That doesn't mean that. That is that there won't be a need for professional translators anyway, in the future.

Igor Afanasayev 29:42
Okay. All right. Thank you so much. Again, we we got lots more questions that we can answer live. They will be posted in the community and we will in the IQ test there so you can answer them at your own pace, and share your presentation as well. All right now we will be starting a poll that will help you understand how, how people are thinking about the presentation overall about the questions, etc. I think we'll be starting right now.

Tess Whitty 30:11
And well, they will they tag me in the community, so I will see it or,

Igor Afanasayev 30:16
yeah, I believe we as organizers of the conference, we will invite you to the right, I will like pre create the right thread for you in our community. So you

Tess Whitty 30:25
can do I pick a winner? Or do you do that? Yeah, yeah, absolutely.

Igor Afanasayev 30:29
Absolutely. Well, we are waiting for the poll to be to be launched. What would be the best question that you had a chance to answer today so far?

Tess Whitty 30:42
Oh, it's I'm saying the same as an announcer This is really hard to pick and I think I only answered about three. So I'm gonna pick the question from the trainee beginner translator. Because I think that he could benefit a lot from the book. Okay. Oh, that was? Who was that from?

Igor Afanasayev 31:08
Yeah, we'll find that answer that question from the list of Oh, find the person. All right. Thank you so much again. We really enjoyed you being with us today. And with that, we will be starting a poll.

Tess Whitty 31:24
Thank you for having me here.

Igor Afanasayev 31:26
Thank you.