Foreign alphabet soup (image generated by AI)

AI Chat Support for Foreign Language Alphabets

I turn to AI first and foremost for content creation, as it’s so good at creating model foreign language texts. But it’s also a pretty good conversational tool for language learners.

That said, one of the biggest obstacles to using LLMs like ChatGPT for conversational practice can be an unfamiliar script. Ask it to speak Arabic, and you’ll get lots of Arabic script. It’s usually smart enough to work out if you’re typing back using Latin characters, but it’ll likely continue to speak in script.

Now, it’s easy enough to ask your AI platform of choice to transliterate everything into Latin characters, and expect the same from you – simply instruct it to do so in your prompts. But blanket transliteration won’t help your development of native reading and writing skills. There’s a much better best of both worlds way that does.

Best of Both Worlds AI Chat Prompt

This prompt sets up a basic conversation environment. The clincher is that is give you the option to write in script  or not. And if not, you’ll get what script should look like modelled right back at you. It’s a great way to jump into conversation practice even before you’re comfortable switching keyboard layouts.

You are a Modern Greek language teacher, and you are helping me to develop my conversational skills in the language at level A2 (CEFR). Always keep the language short and simple at the given level, and always keep the conversation going with follow-up questions.

I will often type in transliterated Latin script, as I am still learning the target language alphabet. Rewrite all of my responses correctly in the target language script with any necessary grammatical corrections.

Similarly, write all of your own responses both in the target language script and also a transliteration in Latin characters. For instance,

Καλημέρα σου!
Kaliméra sou!

Do NOT give any English translations – the only support for me will be transliterations of the target language.

Let’s start off the conversation by talking about the weather.

This prompt worked pretty reliably in ChatGPT-4, Claude, Copilot, and Gemini. The first two were very strong; the latter two occasionally forget the don’t translate! instruction, but otherwise, script support – the name of the game here – was good throughout.

Try changing the language (top) and topic (bottom) to see what it comes up with!

 

An illustration of a cute robot looking at a watch, surrounded by clocks, illustrating AI time-out

Avoiding Time-Out with Longer AI Content

If you’re using AI platforms to create longer language learning content, you’ll have hit the time-out problem at some point.

The issue is that large language models like ChatGPT and Bard use a lot of computing power at scale. To keep things to a sensible minimum, output limits are in place. And although they’re often generous, even on free platforms, they can fall short for many kinds of language learning content.

Multi-part worksheets and graded reader style stories are a case in point. They can stretch to several pages of print, far beyond most platform cut-offs. Some platforms (Microsoft Copilot, for instance) will just stop mid-sentence before a task is complete. Others may display a generation error. Very few will happily continue generating a lengthy text to the end.

You can get round it in many cases by simply stating “continue“. But that’s frustrating at best. And at worst, it doesn’t work at all; it may ignore the last cut-off sentence, or lose its thread entirely. I’ve had times when a quirky Bing insists it’s finished, and refuses, like a surly tot, to pick up where it left off.

Avoiding Time-Out with Sectioning

Fortunately, there’s a pretty easy fix. Simply specify in your prompt that the output should be section by section. For example, take this prompt, reproducing the popular graded reader style of language learning text but without the length limits:

You are a language tutor and content creator, who writes completely original and exciting graded reader stories for learners of all levels. Your stories are expertly crafted to include high-frequency vocabulary and structures that the learner can incorporate into their own repertoire.

As the stories can be quite long, you output them one chapter at a time, prompting me to continue with the next chapter each time. Each 500-word chapter is followed by a short glossary of key vocabulary, and a short comprehension quiz. Each story should have five or six chapters, and have a well-rounded conclusion. The stories should include plenty of dialogue as well as prose, to model spoken language.

With that in mind, write me a story for French beginner learners (A1 on the CEFR scale) set in a dystopian future.

By sectioning, you avoid time-out. Now, you can produce some really substantial learning texts without having to prod and poke your AI to distraction!

There may even be an added benefit. I’ve noticed that the quality of texts output by section may even be slightly higher than with all-at-once content. Perhaps this is connected to recent findings that instructing AI to thing step by step, and break things down, improves results.

If there is a downside, it’s simply that sectioned output with take up more conversational turns. Instead of one reply ‘turn’, you’re getting lots of them. This eats into your per-conversation or per-hour allocation on ChatGPT Plus and Bing, for example. But the quality boost is worth it, I think.

Has the section by section trick improved your language learning content? Let us know your experiences in the comments!

Parallel text style learning, like Assimil courses, can be a great way to improve your fluency.

DIY Assimil : Parallel Text Learning with ChatGPT

Assimil language learning books are hugely popular in our polyglot community. And for good reason – many of us learn really effectively with its parallel text method.

They’re especially userful when the base language is another of our stronger languages, adding an element of triangulation. I learned a heap of Greek vocabulary from the French edition Le Grec sans Peine, at the same time as strengthening my (ever slightly wobbly) French.

Now, Assimil is already available in a great range of language pairs. But it’s not always a perfect fit. For example, some editions are more up-to-date than others. More off-the-beaten-track languages still aren’t available. And at times, you can’t find the right base language – no use learning Breton through French, if you don’t have any French.

Enter ChatGPT (or your alternative LLM of choiceBing also does a great job of these!).

DIY Assimil Prompting

Copy and paste this into your AI chat, changing the language (top), translation language (middle) and topic (bottom) to suit.

You are an expert creator of language learning resources. I want to create some text-based learning units for beginner Malay learners (level A0/A1 on the CEFR scale). The units follow the parallel text approach of the well-known Assimil language learning books.

Each unit has a text in the target language (about 250 words) on a specific vocabulary topic. It should be narrative, talking about how the topic relates to an everyday person. It should be divided into logical paragraphs. After each paragraph, there is an English translation of that paragraph in italics.

The text should be written in very clear, simple language. The language must read like a native speaker wrote it, and be error-free and natural-sounding. Source the info for the text from target language resources online, making it as up-to-date and authentic as possible. It should be completely original and not copied or lifted from any other source directly.

After the text, there is a glossary list of the key topic words from the text, sorted alphabetically and grouped by parts of speech (nouns, verbs, adjectives, adverbs etc.).

Are you ready to create some content? The first topic is: Mobile Technology

This prompt creates a prose-based parallel text unit. However, if you prefer dialogue-style texts, simply change the second paragraph of the prompt:

Each unit has a humorous dialogue in the target language (about 20 lines) on a specific vocabulary topic. The dialogue should relate the topic to everyday speakers through colloquial, idiomatic language.

The prompt works a treat in both ChatGPT Plus (paid) and Microsoft Bing (free). I also got very useable results in the free version of ChatGPT and Claude 2. It works so well as the focus is purely on what LLMs do best: spooling off creative text.

How Do I Use Them?

So, with your shiny, new Assimil-style units spooled off, what do you do with them?

Personally, I like to copy and paste the output into the notes app on my phone. That way, they make nice potted units to browse through when I have some spare moments on the bus or train. They’re equally handy copy-pasted into PDF documents that you can annotate on your phone or tablet.

Parallel text for Malay language learning created by AI

Parallel text in Malay and English created by AI

In terms of real-world use, the self-contained, chatty texts typically created make perfect material for the islands approach to improving spoken fluency. Create some units in topics that are likely to come up in conversation. Then, spend some time memorising the phrases by heart. You’ll be able to draw on them whenever you need in real-life conversation.

Enjoy prompts like these? Check out my book AI for Language Learners, which lists even more fun ways to get results without paying hefty course book price tags!

Fun With Texts : Travel Edition

I came across an ancient video this week that took me right back. The video in question  was from a series of video diary entries I made on a trip to Austria in 2004. In this particular segment, I was proudly showing off the stash of free leaflets I’d cached from Klagenfurt town hall – treasures of authentic texts to take home for my teaching materials box.

German-language texts from Austria - leaflets about the EU in 2004

Austrian leaflets about the EU (2004)

A still of Rich West-Soley showing some leaflets from Austria in a video from 2004

Showing off my Austrian leaflet haul in 2004 in a video shot on a phone just a little more sophisticated than a toaster, judging from the quality

Fast forward 19 years, and I’m approaching the end of a wonderful, extended trip around Greece. It’s been a holiday full of wonderful sights, amazing food, and of course, lots of language practice. Incidentally, Greeks must be amongst the most encouraging people on the planet when you try to speak their language.

But what links this trip with that early noughties vid is that continued fascination with curating authentic texts. It’s a polyglot obsession that’s lasted well beyond my classroom teaching days; there’s no longer any teaching materials box to fill, but I’m still on the hunt.

Hunting Texts : Then and Now

The format has changed, naturally. It’s less about free brochures and leaflets now. Alas, my EasyJet baggage allowance won’t quite stretch to that any more. This time, it’s digital – and I’ve been going to town collecting text samples for my virtual Greek learning box.

Of course, Greece has a tradition of texts that stretches back a little further than many fellow European countries. It’s been particularly fun looking out for inscriptions on the many ancient monuments, and spotting similarities and differences between the ancient and modern languages.

A stone tablet in an Ancient Greek ruin, with a partial inscription in Greek

Authentic Texts in stone!

An Ancient Greek artefact


But it’s the modern examples that really hit the spot – the more everyday and prosaic the better. From bags of crisps to public notices, every bit of writing is a potential new word learnt, and an extra peep into the target language culture. It’s addictive.

A notice to save water on a Greek ship

Save water, and save those words (in Anki!)

A washing machine control panel with Greek labelling

It’ll all come out in the wash

As far as I’m concerned, there’s never any going over-the-top when collecting digital texts. Knock yourself out with as much target language as you can! The criteria for what makes an authentic text are wildly broad – it can be the odd couple of words, a text-dense poster or an entire book. It all has worth to us as learners, no matter how long.

The only rule I try to stick to is one of practical use; I aim to try and use the images somewhere, be it a blog post (like this one on German political posters) or by scraping the language for Anki flashcard entries.

A bag of crisps with Greek labelling

Language snacks

Are you a curator of authentic texts in your target language? How do you collect them, and what do you do with them afterwards? Let us know in the comments!

Nigheanan Mòra by Catrìona Lexy Chaimbeul (2014). Reading target language texts is an excellent way to improve foreign language skills.

Working with Target Language Texts

Eager to push my Gaelic out of the language course box and into the wild, I’ve been working with a number of short texts for intermediate learners lately. Luckily, quite a few readers have appeared in the recent years, including a bunch of fun titles that go beyond the usual ‘Celtic myths retold’ route (not taking anything away from the great series of beginners’ books from Jason Bond).

A recent favourite of mine, Nigheanan Mòra (Big Girls, 2014), was penned by one of the creatives behind recent BBC Alba drama hit An Clò Mòr, Catrìona Lexy Chaimbeul. On the surface, it’s in firm rom com territory, at turns silly, funny and melodramatic. But it’s grown-up enough to feel like you’re reading a real book, and not just an oversimplified, fleshless yarn that trades plot for easy reading. It’s also chock full of colloquial, conversational Gaelic dialogue, which makes for a great living language learning model.

That said, getting the most from a reader takes a bit more organisation than simply starting at page one and ploughing through. Better to have a strategy to maximise both your enjoyment and your learning.

Working With Texts : One Approach

Of course, there’s no single ‘correct’ way to work through target language texts. Through trial and error, I’ve found a way that works for me, which I’ll outline here. It works best with short-ish texts, since it involves two passes in quick-ish succession, but you could also use it with short sections of longer texts.

That’s because manageable chunk size is the key to this method. Often, you won’t need to worry about that with texts specifically for learners. Many books that support learners, like Nigheanan Mòra, already have nice short chapters of 5-10 pages. I find that’s the ideal length to read and digest texts without tiring (because, let’s face it, reading in a foreign language is more taxing). If chapters are much longer, just flick ahead a little way to see if there’s a natural stop somewhere, and make that your goal.

Pass One : the Chill read

After that, it’s time to start reading. The first pass is the no-chill literary gambol. Read for gist and plot, and don’t fret a jot about the odd unknown word. The focus here is on simply understanding and enjoying the story, first and foremost. I like to go full non-study mode at this stage. I’ll pick a cosy reading spot, grab a drink and just try to immerse myself in the story. No dictionaries, no pencils, no interruptions.

After that first reading – maybe 20 minutes or so – I’ll stop, take a breath, and reflect on the twists and turns of the plot. It’s important to take a passive break to cogitate calmly like this, given that our brains work more efficiently with pacing (a trait the Pomodoro technique plays into).

Pass Two : The Close Read

After this brief pause, I’ll then flick back casually through the pages I’ve just read. In particular, I’ll revisit those passages I felt were tough, or noticed myself slow down in during the first pass. For each one, I’ll re-read carefully, this time trying to translate in my head, paying more attention to the grammatical structures. I’ll also spend some time on words I didn’t get the first time round, looking for contextual clues to help guess the meaning (and not reaching straight for the dictionary).

This is the stage where I really prefer old-school paper books to Kindle ones. I’ll have a pencil by me, underlining any turns of phrase that sound really idiomatic or conversationally useful. With a pop-story like Nigheanan Mòra, there’ll be loads of those, thanks to all the snappy dialogue.  They’re the snippets where I’ve realised aha! So that’s how you say X in Gaelic.

Finally, after all that, I’ll spend some time cross-referencing those new structures in grammars and online materials like the LearnGaelic.scot dictionary and Wiktionary. Once I’m sure I’ve understood them, I’ll add the phrases to my Anki deck. Adding phrases is so much more effective that lifting just individual words from texts. We speak in phrases, not lone words, so by the end of this stage I have some truly useful material to drill. This phrase-lifting approach thoroughly mines a text for connectives and sentence frames – the bread and butter of fluency.

Find What Works For You

So there you have it – one way to work with authentic texts. It’s not rocket science or particularly groundbreaking, but it works for me. And it helps, in terms of discipline, to know that I have these regular steps to follow, to give my target language reading some kind of structure.

What I also find invaluable about it, in terms of motivation, is building in a reading for pleasure stage, which includes choosing material I find fun, as well as the time to enjoy it without pressure. Even if that is silly old rom coms.

After all, learning and practising languages shouldn’t just be work, work, work.

There are myriad ways to approach target language texts. What works for you? Let us know in the comments!

ChatGPT writing a short story in German.

Short Stories… in ChatGPT

It’s no secret – reading fiction is a favourite strategy of polyglot learners. That’s more than simply reading Harry Potter novels in translation. There’s a whole market sector that revolves around non-native short stories, and I’m not alone in enjoying the excellent Short Stories In… or Penguin Parallel Texts series to practise my languages.

But what if we could source those stories on demand… and for free?

Unless you’ve been hiding for the past three months, you’ll know where I’m going with this. ChatGPT, the natural language processor, has already made ripples in the fan fiction arena. And, it turns out, it has a knack for performing the same feat multilingually, and tailored to your exact needs.

The power of it becomes apparent when you ask it to write you a story. Because you can tailor that story precisely to your own interests. Personal interest, of course, is a holy grail with language learning motivation. And ChatGPT is like your own private author, ready to fit original content to exactly what you like.

I started where I started – literally, with languages – and requested a German short story about Eurovision. What else? The results were pretty impressive.

ChatGPT writing a short story in German.

ChatGPT writing a short story in German.

The only thing is, it’s a bit wordy for my (hypothetical) class of German students. So I ask ChatGPT to tailor it to a specific level:

ChatGPT writing a short story in German.

Tailoring the story to a specific level.

Brilliant – we’re getting something we can turn into a learning resource now. But I’d love my students to focus on more descriptive adjectives to improve their writing. Can we turn this into a better model?

ChatGPT writing a short story in German.

Tweaking the output with specific criteria.

Again, ChatGPT turns up the goods! The German is sound, and the story is a fun little read. But what about making this a polyglot resource, parallel resource, so anyone learning more than one language can keep their learning in sync? No problem:

ChatGPT writing a short story in French.

Translation into French.

Impressive. It has no issue with any of what you’d call the mainstream languages. I tried it in all of the languages I have some proficiency in, and it even churns out decent Greek and Polish. I’m not yet fluent enough in Scottish Gaelic to check this properly, but it seemed the only one that was a bit iffy, despite giving it a good go:

ChatGPT writing a short story in German.

A translation into Scottish Gaelic.

Finally, let’s throw in a short summary version we can use as revision materials, or an item description:

ChatGPT writing a short story in French.

A short summary of the story in French.

Obviously, this all comes with the caveat that it needs careful checking before use as an accurate resource. But the initial performance is pretty spectacular, to be honest. As the model is tweaked and improved, it’s not hard to imagine this becoming a cornerstone of personal resource creation for learners of languages, as well as everything else.

Magic of the Mundane : Language Learning from the Inbox Clutter

We language learning enthusiasts can turn the most mundane, dull items into shiny, valuable objects of curiosity and enrichment – much like cats and dogs frequently manage. It’s a very special gift we have.

I’m not saying we can play for hours on end with a cardboard box or some wrapping paper. I’m talking about the mundanity of digital life, particularly those parts of it which normally leave us a little fuming.

Take electronic newsletters. Yes, those all-too-frequent, clog-up-your-inbox ad mail-shots from companies, websites and other organisations you (usually) provided with your email address in weaker moments. If your inbox resembles mine in any way, you probably have more of this automated, targeted (but totally solicited) junk than emails from actual human beings.

Predictably, when I get these in English, my reaction usually ranges from blasé curiosity and a quick skim through, to mild annoyance and immediate deletion.

But add a foreign language to the mix, and they’re magically transformed. They’re now language learning resources ™️! <cue amazed oohs and aahs>

From the Mundane to the Sublime

It’s a magic trick that can add appeal to the most prosaic of inbox items. This week, I found myself transfixed by an email ad for branded pots and pans from my favourite Greek TV chef, Akis. Do I have any interest in cooking? That’s debatable. But is it a whole lot of geekish fun learning words for specialist kitchen utensils in Greek? You bet.

From the mundane to the sublime : an image promoting Greek chef Akis Petretzikis’ range of pots and pans.
From the mundane to the sublime : an image promoting Greek chef Akis Petretzikis’ range of pots and pans.

If you’re looking to find the value in the e-marketing chaff, it’s easy enough to seek out these kind of target language bulletins nowadays. Call up the websites of your favourite brands or personalities, add your details, and click submit. Instant brand servitude with a dash of language learning thrown in. A bit of pop culture surfing doesn’t hurt, either.

But just to avoid over-gorging on that language learning feast and passing out from junkbox marketing fatigue, consider using email rules to siphon them off to a special folder (or even a dedicated email just for target language newsletter clutter). It’s an enthusiasm-saver if you end up signing up for a couple too many. And we’ve all had too much at the buffet before, after all.

Learning Places : Banff and Macduff Bay

Learning Places

Do you ever find yourself learning the same things in the same places? Not intentionally, but somehow drifting into the same languages, the same subjects, according to where you are?

I’ve been on a well overdue parental visit on the north-east coast again this week after a hectic few months. And I’ve noticed that something of a pattern has emerged. Despite having a gloriously diverse language library to pick and choose from, and well-made general dabbling intentions, I always go for a Gaelic book while I’m staying here.

Ambient Inspiration

Maybe it’s seeing the mountains of East Sutherland far across the water, near which Nancy Dorian researched the sadly moribund local dialects of Gaelic, that prompts me. Or perhaps it’s just that quintessentially Scottish coastal mix of sea, sky and moorland that gets me in the mood. Whatever it is, I’m more likely to settle down with Gaelic here than in any of my other favourite reading places.

Places for learning: Banff and Macduff Bay

Banff and Macduff Bay, across from which the hills of Sutherland can just be made out

There’s a potential downside to that, according to some educational psychologists. Research into context-dependent memory suggests that material learnt exclusively in one environment may be harder to recall in others.

But the flip-side of that is the use of wider memories as triggers for learnt material. In my case, think of home, remember your Gaelic.

The choice of book says a lot, too. As if to match the homely nostalgia, on each trip home I’ve been working my way further into Roderick Mackinnon’s original Teach Yourself Gaelic, 1971 (1992 reissue). A treasure of an old language book, if ever there was one.

Learning places : A copy of Teach Yourself Gaelic in front of a seaside view

An unashamedly posed shot of Teach Yourself Gaelic

 

What languages, books and places go together for you? Do you have a preferred nook or corner of the world to settle down to some study in? Let us know!

Greek microblog content from Instagram (screenshot).

The Way of the Microblog : Kitchen Sink Inspiration and Language Learning

It’s all about the foreign language microblog for me lately. Short, snappy snippets of target language piped directly to your social media streams: what’s not to love?

In fact, I’m practically drowning in them at the moment. That’s thanks to the notorious and mysterious algorithm (TM), of course, which is a fact of life these days; like one thing, and you get a ton more of the same thrown at you, for better or for worse.

Happily, in the case of us language learners, it’s generally for the better. Take my Instagram feed; its AI wisdom has decided to channel reams of Greek pop psych, heartwarming quotes and concise self help my way. It’s twee and a wee bit naff, granted. But every one of those posts is a 30-second language lesson.

This latest bite-sized adventure all started with a single Greek account, gnwmika.gr. It exclusively posts what you might call ‘fridge magnet’ content: folk wisdom and kitchen sink inspiration.

The great lesson imparted here, in true, lofty microblog style, is:

“Beautiful things will make you love life. Difficult ones will teach you to appreciate and respect the beautiful ones.”

I know – deep, eh.

Anyway, I hit follow and thought little else of it… Until things escalated. Next thing, I’m being shepherded to not only more of the same, but anything and everything Greek. Poetry, history, celebs, TV… the lot. It’s become a rabbit hole leading to some well obscure (but fascinating) places. And, crucially:

…my Greek is so much better for it!

Fill Your Little (Microblog) World Right Up

It all plays in marvellously to the fill your world with target language strategy. Since our worlds are ever more digital, one of the easiest ways to do that is to follow the monkeys out of accounts we find fun and engaging. Add one or two, and let the system start popping more and more into your suggested follows.

Now, the only catch is that the algorithm (TM) is smothering me in Greek. I’d love a bit of Gaelic, Icelandic, Norwegian or Polish (and the rest). So, if you’re reading this and have some good microblog recommendations to kick the cycle off again…

…please let me know!

A wee book treat to myself: Colloquial Scottish Gaelic (Routledge)

A Book in the Hand (Is Worth Two in the Kindle Library)

Sometimes I forget how much I love to hold a real book in my hands.

Don’t get me wrong. I love the convenience of Kindle titles and other e-formats. Only the other week I was singing the praises of the Teach Yourself enhanced versions. A whole course – text and audio – in a single place (and it adds 0kg to my backpack weight). I still think they’re fantastic.

But sometimes you get a reminder of how satisfying old school is. I had one this week when I finally plumped for a long yearned-for hard copy of Colloquial Scottish Gaelic.

Why had I put it off for so long?

Well, there’s the price of the hard copy, for a start. £35 is a hefty commitment for a book. Especially so, considering that I had access to the electronic version for free through my university library. Not only that, but like many publishing platforms making audio content free, Routledge has put all the audio online. I could access all of the content already!

But for all that, I just wasn’t bothering to use the materials at all. Why? screams the spendthrift inside me.

Fast forward, my Amazon credit spent, and the book proudly on my shelf. I’m picking it up at every opportunity, having a quick nose here and there when I notice it, sitting down for half an hour’s mooch through the pages. I’m even listening to those audio materials and reading along, finally.

So what is so different?

It’s hard to put your finger on just what is so special about a real book. There’s the joy of the tangible ownership of it, perhaps. I made an investment in a thing – now I want to make the most of that thing. It’s almost like you can feel the weight of the knowledge you’ve paid for right in there.

And there’s nothing like using money (or vouchers) to feel the value of a physical object. I admit I get a bit of that as I curate a Kindle library. It’s lovely seeing the digital books line up neatly on those shelves.

But there’s  something simply cosy (or hyggelig, or gemütlich etc.) about holding a real book in your hands, isn’t there?

And sometimes it takes a wee treat to yourself to remind you of that.