Archives for February 2023

Will ChatGPT revolutionise machine translation?

24.2.2023 by Jari Herrgård

ChatGPT was launched in November 2022, and it feels like the world – or at least the future of information work – has not been the same since.

ChatGPT is the first service that makes an artificial intelligence based on a large language model (LLM) available to the everyday user and provides an easy user interface for it. Even in this case, artificial intelligence isn’t an actual “intelligence” but a mathematical model that calculates which words are likely to follow one another. However, ChatGPT does this more smoothly and with better versatility than the previous models.

At best, ChatGPT improves human productivity

At first, it may seem that the model can do anything: it types code, writes snappy articles and responds to questions. Naturally, it also does translation! In ChatGPT’s user interface, it is possible to request translations and other texts using natural language prompts in the same way as when asking from another person. I asked it for a poem on selling localisation services, and it even managed to do that. In terms of style, the output was much better than what I myself would have been able to produce.

However, when you look at texts produced by ChatGPT a little more closely, you will see that it suffers from the same issues that plague machine translations by proficient neural networks, for example – the translations are often correct but usually only almost correct, and the AI may come up with sources that it cites as if scientific research.

Due to the errors it makes, ChatGPT is, in a way, very human. Regardless, this does not make the quality of its text poor in general but rather reminds us of its limitations. If you want to make use of the output of an LLM, it would be a good idea to have a human check it or, at the very least, you should acquaint yourself with what kind of text it produces in a specific context.

If you remember and are mindful of these limitations, ChatGPT is certainly a useful tool in various industries. When used properly, it increases human productivity by coming up with ideas, checking and making drafts.

What about translation?

ChatGPT is surprisingly adept at translation. In the chat user interface, translation is easy to do by using free-form requests. What separates ChatGPT from traditional machine translation tools is that you can also give it instructions for the translation task. Below are a couple examples of this. In the first, I requested a translation in the style of a pirate ( what could be more useful?), and in the second, a couple of alternative translations.

And it can also translate into several languages of your choosing with just one prompt:

So, producing translations is easy, and they seem fairly correct as well (while my language skills aren’t sufficient for assessing all of the examples). In some cases, it even adds comments that explain its decision-making process.

Thus far, there have only been preliminary evaluations on the accuracy of the translations produced by ChatGPT compared to the prevailing neural machine translations (NMT), and it appears that the translation performance of LLMs cannot compete with the more specialised machine translation tools. The difference may not be huge, but it is still significant enough to make a difference in professional settings. However, there have not been many comparative studies yet, so we should not be too quick to draw further conclusions. In any case, it seems that ChatGPT beats the neural network models in specific context recognition or memory tasks.

LLMs in professional use

The entire translation industry has now turned its gaze to large language models, wondering if they will replace NMT-based machine translation tools or even translators themselves. Talk of the eventual redundancy of translators is nothing new, but that hasn’t happened so far – and I don’t think it will happen now. But LLMs will undoubtedly have some sort of impact.

The examples below are simple and easy, but even they show that you get lots of possibilities for translation by using different prompts compared to traditional machine translation tools. The current machine translation tools always produce one version in a set way. You cannot influence their translation output except by training the tools on materials of your choosing or by using an adaptive system and feeding it with texts in advance.

Through the use of prompts, ChatGPT can make adjustments on the fly, which you could make use of with the help of example texts or translation memories, for example. In the same way, it also offers possibilities for editing the ready text to follow a certain style.

As ChatGPT supports well the production of entire texts, it may shift part of the translation work away from traditional translation, thus moving the focus more on creating a new text in another language. Of course, this is what translation basically is already, but due to practical reasons, translations are often very closely tied to their source language. In time, machine-assisted methods used in the translation services industry could more clearly be divided by their intended use. This means the method more closely linked to the source language could be retained in technical texts but would give way in more creative translations.

Despite the hurdles, there are endless possibilities

LLMs will inevitably become part of additional services in translation. For example, in term base work, ChatGPT can already be successfully used for rough work on short texts. Perhaps there are more uses in text analysis as well? There must be many more uses that I do not even realise or which are not even a thing yet.

So, there may be endless possibilities, but there are also many hurdles. At the moment, one of the most significant hindrances is data security: ChatGPT cannot be used privately, which prevents its use in many situations. However, I expect services with better data security and privacy to find their way to the market soon. When it comes to translation tools, accuracy also plays a crucial role, and as long as LLMs cannot compete with neural network-based models, they will have less use in professional translation.

Currently, there are more questions than answers regarding the use of LLMs in the language services industry, and no one can say for sure where these models will be used. In the coming years, LLMs and ChatGPT are certain to shape the language services industry in a decisive way and create new ways of working and new competence areas, starting from what kinds of prompts users know to use when requesting texts from AIs.

My guess is that we will have new important LLM-based machine translation tools and text-production solutions, the mastery of which will (for a time) become a competitive edge. And even this tool is used by humans, whose productivity it may increase a lot.

Language tip of the month: 3 + 1 tips for polite English

22.2.2023 by Delingua

Brits in particular but also other native English speakers like to be as polite as possible, whereas Finns tend to say things far more directly.

Whether it’s an email or communicating on the phone or face to face, Finns handle situations by getting right down to business. Even if this kind of communication is ordinary in Finland, it may sound rude to an English speaker. So if you want to avoid blunt expressions and aim for fluent English, start using these tips for polite English!

Please

Have you ever paid attention to the use of please in English? Please is an important word that you should keep in mind in both face-to-face interaction and written communications, and forgetting it is actually somewhat rude.

Over and over again, I’ve heard Finns ordering a pint at the pub with the clunky phrase “I take a beer”, when the native way to politely express it would be “Could I have a beer, please?”. In Finland, it’s business as usual in a café to say ”Ottaisin yhden kahvin.” (I’d have a coffee), whereas for an English speaker, the appropriate phrase would be “Could I have a coffee, please?”.

In Finnish, using the word kiitos (thank you) is common, but there is no direct equivalent for the word please, which means the same level of politeness is expressed with the Finnish grammatical conditional -isi-. This is why Finns often think it is sufficiently polite to use the word could or would, but if it is a request and not a question, it definitely also requires a please. Saying “Could you pass me the salt?” and “Could you pass me the salt, please?” are entirely different matters.Adding please shows your appreciation.

It’s good to keep in mind that the word order is also significant. You can add the word please to either the start or the end of the phrase, depending on if you intend to say it more as a request or as a demand. At the end of a phrase, the meaning is very similar to the word kiitos in Finnish. If it is at the beginning of a phrase, it lends more weight to the word. The phrase “please behave” instantly has a more demanding, commanding tone.

Sorry and excuse me

In English, apologetic words are thrown around fairly casually. In Finnish, apologies are only made when they are genuinely meant. In English, the words sorry, excuse me, pardon and apologies are used constantly in everyday language. Sorry and I’m sorry are a good fit in almost any situation, whether it’s bumping into someone in a crowded bus or a genuine, heartfelt apology. When you can’t hear the other person, the correct phrase is “Sorry, I didn’t catch that” or “Pardon?”.

The expression excuse me is used when you want to interrupt or catch someone’s attention. It’s also used to express that you are leaving the scene. The phrase is used, for example, in the following phrases: ”Excuse me, could you tell me where I might find Mr Smith?”, “Excuse me, may I ask your name again?” or “Excuse me, I need to take this phone call.”.

However, you have to be on your toes with excuse me, because with the wrong tone of voice, it can express sarcasm or even be aggressive. ”Excuse me, but I wasn’t born yesterday!”

The phrase my apologies is used in more official connections, such as in written communications when expressing regret or turning down an invitation, such as in ”My apologies for this oversight on our part.”.

Ole hyvä is not please

Whereas Finnish has the one and only ole hyvä, English has dozens of equivalents for it, depending on the use. If someone says thank you, you usually respond by saying you’re welcome but also phrases like my pleasure and not a bit are equally suitable, along with many other similar phrases.

Sometimes you can hear Finns erroneously saying please when giving someone something. English has its own phrases for this situation: here you are or there you go. When you want to give someone space or indicate that it’s their turn to speak, you can use, for example, the phrases be my guest or the floor is yours.

General politeness

In written communications, it’s important to pay attention to how you formulate your message. The Finnish style is to say everything fairly bluntly. When giving feedback in Finnish, you can easily say something that corresponds to “We would like a longer term of payment” or even “We want the term of payment to be 30 days.” In English, you should always soften the message and say, for instance, ”Could you please consider amending the payment terms to 30 days?”

Did you like what you read? Did you perhaps learn something new? In the Language Tip of the Month series, we are giving language tips on a variety of topics. Follow us on social media so you don’t miss the next article!

Translation profitability (ROI)

9.2.2023 by Katja Virtanen

Nowadays, new content is produced constantly. At the same time, there is an increasing need for translation and localisation, i.e. the adaptation of content into different languages and cultures. When even global publishing is fast and easy, the question that comes up is which languages the content should be translated into and in which languages the different language versions should be maintained.

And how might return on investment (ROI) be measured when it comes to translations and the maintenance of language versions?

Objectives of buying translations

The buying of translations should always be linked with business objectives.

First, consider why software or materials, for instance, should be translated into a certain language and what are the objectives that are pursued with the language versions.

Most commonly, the aim of adding a new language version is to attract more customers and sales. Translations are needed because studies indicate that people prefer to make purchases using their own language. If increasing net revenue or the number of customers or users in a language area is a strategic goal for you, follow the development of these figures after adding the language version and compare them with the corresponding figures before the language was added.

Then again, things that are more difficult to measure, such as brand recognition or customer satisfaction in a language area, can also be interesting indicators.

Therefore, the business benefits of buying translations are best calculated on the basis of what is most important for the business.

Calculate (all) translation costs

Translation prices are often given on the basis of the source text word count. That makes it easy to calculate how much translations cost. Therefore, the customer is often very tempted to choose the cheapest service provider.

It is more difficult to calculate the hidden costs of translations, such as the cost of the internal administrative work related to ordering translations or the time spent correcting poor translations. If translations are ordered from different providers, costs are also incurred by translating the same materials or parts of them several times.

Moreover, it should be remembered that buying translations is rarely a one-off expense item: the materials must be updated regularly and websites in different languages often also lead to costs related to foreign-language customer service and marketing, for instance.

Think about profit as well

If calculating the actual translation costs is challenging, it is even more difficult to calculate translation profitability. In this context, the language service research company Common Sense Advisory has found that the vast majority of companies that spend substantial amounts of money each year on translations measure only costs and not the opportunities created by translations at all.

If you want to measure monetary profit, the actual ROI is obtained by dividing the net profit generated by the net revenue from the language area by the investment costs. Naturally, ROI is not a static figure, but should be reviewed at regular intervals.

ROI only takes monetary results into account, although other important objectives can also be achieved through translations. For instance, brand recognition and customer loyalty can be a result of well-localised language versions and are extremely important for a company’s success.

Profit indicators

Since translation with all its related costs requires an investment, the benefits yielded by the investment should be measured in one way or another. However, the potential return on investment can be challenging to measure as a percentage. It is particularly difficult to assess what is ultimately due to the language version and what is due to other efforts made. Even if there isn’t really a perfect formula for calculating the ROI for translations, localisation is an enabler for many other internationalisation-related matters that can be measured.

In addition to the growth indicators mentioned above, good indicators in the digital era include, for instance, website traffic volumes, visitor engagement or the conversion rate. For instance, you can count website impressions and the number of downloads or compare the success of a localised advertisement campaign in different language areas. If there is a language area where nothing seems to work, consider if something could be done and whether it’s worth maintaining the language version in the first place.

Improving translation profitability

The higher the ROI of translations, the better the cost control and the higher the return on the investment. Here are a few tips on saving costs and improving profitability:

1. Concentrate translations

Choose a good translation agency and concentrate your translation assignments to a single place. This ensures your translation partner learns your special terminology, quality remains consistent and you can benefit from seamless cooperation optimally. Smooth cooperation reduces your own administrative work.

2. Take advantage of translation technology

Professional translation agencies use technology that makes your life easier and also generates clear savings. Make the most of translation memories, various integrations with your own systems and the opportunities offered by machine translation. Accelerating the translation process will increase profitability.

3. Add languages and keep an eye on profit

Adding language versions has never been this easy. Translating a single website into one language is a relatively small investment, but the benefits may be limited. So, choose the languages carefully and keep an eye on the performance of different language versions. Add and remove language versions as necessary.

If you need more tips on how to measure translation profitability, read our guide on the topic or contact us! We are happy to help improve the profitability of your translations.