ChatGPT takes conversation to the next level with Advanced Voice Mode

ChatGPT Advanced Voice Mode is Finally Here (For Most of Us!)

Finally – and it has taken SO much longer to get it this side of the Pond – Advanced Voice Mode has popped up in my ChatGPT. And it’s a bit of a mind-blower to say the least.

Multilingually speaking, it’s a huge step up for the platform. For a start, its non-English accents are hugely improved – no longer French or German with an American twang. Furthermore, user language detection seems more reliable, too. Open it up, initiate a conversation in your target language, and it’s ready to go without further fiddling.

But it’s the flexibility and emotiveness of those voices which is the real game-changer. There’s real humanity in those voices, now, reminiscent of Hume’s emotionally aware AI voices. As well as emotion, there’s variation in timbre and speed. What that means for learners is that it’s now possible to get it to mimic slow, deliberate speech when you ask that language learning staple “can you repeat that more slowly, please?”. It makes for a much more adaptive digital conversation partner.

Likewise – and rather incredibly – it’s possible to simulate a whole range of regional accents. I asked for Austrian German, and believe me, it is UNCANNILY good. Granted, it did occasionally verge on parody, but as a general impression, it’s shocking how close it gets. It’s a great way to prepare for speaking your target language with real people, who use real, regionally marked speech.

Advanced Voice Mode, together with its recently added ability to remember details from past conversations (previously achievable only via a hack), is turning ChatGPT into a much cannier language learning assistant. It was certainly worth the wait. And for linguaphiles, it’ll be fascinating to see how it continues to develop as an intelligent conversationalist from here.

Mapping out conversational probabilities - it's much easier with flowcharts.

Vocabulary Flowcharts : Preparing for Probabilities with ChatGPT

The challenge in preparing for a speaking task in the wild is that you’re dealing with multiple permutations. You ask your carefully prepared question, and you get any one of a number of likely responses back. That, in turn, informs your next question or reply, and another one-of-many comebacks follows.

It’s probability roulette.

What if you could map all of these conversational pathways out, though? Flowcharts have long been the logician’s tool of choice for visualising processes that involve forking choices. Combined with generative AI’s penchant for assembling real-world language, we have a recipe for much more dynamic language prep resources than a traditional vocab list.

And, thanks to a ready-made flowchart plugin for ChatGPT – courtesy of the charting folks at Whimsical.com – it’s really easy to knock one together.

Vocabulary Flowcharts in Minutes

In your ChatGPT account, you’ll need to locate the Whimsical GPT. Then, it’s just a case of detailing the conversational scenario you want to map out. Here’s an example for ‘opening a bank account in Germany’:

Create a flowchart detailing different conversational choices and paths in German for the scenario “Opening a bank account as a non-resident of Germany planning to work there for six months.” Include pathways for any problems that might occur in the process. Ensure all the text reflects formal, conversational German.

The result should be a fairly detailed ‘probability map’ of conversational turns:

A 'vocabulary flowchart' in German, created by the Whimsical.com GPT on ChatGPT.

A ‘vocabulary flowchart’ in German, created by the Whimsical.com GPT on ChatGPT.

Vocabulary flowcharts are another tool in your AI arsenal for speaking prep. Have you given them a whirl yet? Tell us about your own prep in the comments!

A robot interviewing another robot - a great speaking game on ChatGPT!

So Interview Me! Structured Speaking with ChatGPT

The addition of voice chat mode to ChatGPT – soon available even to free users in an impressive, all-new format – opens up tons of possibilities for AI speaking practice. When faced with it for the first time, however, learners can find that it’s all a bit undirected and woolly. To make the most of it for targeted speaking practice, it needs some nudging with prompts.

Since AI crashed into the language learning world, the prompt bank has filled with ways to prime your chatbot for more effective speaking practice and prep. But there’s one activity I’ve been using lately that offers both structure, tailored to your level and topic, and a lot of fun. I call it So Interview Me!, and it involves you playing an esteemed expert on a topic of your choice, with ChatGPT as the prime-time TV interviewer.

So Interview Me!

Here’s an example you can paste into ChatGPT Plus straight away (as text first, then switching to voice mode after the initial response):

Let’s role-play so I can practise my Swedish with you. You play the role of a TV interviewer on a news programme. I play an esteemed expert on the topic of ‘the history of Eurovision’. Conversational turn by turn, interview me in the target language all about the topic. Don’t add any translations or other directions – you play the interviewer and no other role. Wind up the interview after about 15 turns. Keep the language quite simple, around level B1 on the CEFR scale. Are you ready? Start off by introducing me and asking the first question!

The fun of it is that you are the star of the show. You can completely throw yourself into it, interacting with your interviewer with all the gusto and gumption of a true expert. Or you can have some fun with it, throwing it off with silly answers and bending the scenario to your will (maybe you turn out not to be the expert!).

Either way, it’s a brilliant one to wind up and set going before you start the washing up!

A picture of a robot heart - conversation with emotion with Hume.ai

Conversation practice with emotion : Meet Hume.ai

If the socials are anything to go by, so many of us language learners are already using AI platforms for conversation practice – whether text-typed, or spoken with speech-enabled platforms like ChatGPT.

Conversational interaction is something that LLMs – large language models – were created for. In fact, language learning and teaching seem like an uncannily good fit for AI. It’s almost like it was made for us.

But there’s one thing that’s been missing up to now – emotional awareness. In everyday conversation with other humans, we use a range of cues to gauge our speaking partner’s attitude, intentions and general mood. AI – even when using speech recognition and text-to-speech – is flat by comparison. It can only simulate true conversational interplay.

A new LLM is set to change all that. Hume.ai has empathy built-in. It uses vocal cues to determine the probable mindset of the speaker for each utterance. For each input, it selects a set of human emotions, and weights them. For instance, it might decide that what you said was 60% curious, 40% anxious and 20% proud. Then, mirroring that, it replies with an appropriate intonation and flex.

The platform already supports over 50 languages. You can try out a demo in English here, and prepare to be impressed – its guesses can be mind-bogglingly spot-on. Although it’s chiefly for developer access right now, the potential usefulness to language learning is so clear that we should hopefully see the engine popping up in language platforms in the near future!

Scottish Gaelic flash cards with irregular verb paradigms

Going Old School with Language Learning Flash Cards

You might have noticed that I’m partial to a cheat sheet in my language comings and goings. There’s only so much you can hold in short-term memory before a speaking class, and having a scaffold to hand – even gamifying it, where possible – can be a boon. Crib notes, cheat sheets, flash cards – they’re par for the course in language learning. And everyone seems to have their own favourite label for them.

Now, my first thought when making these things is: which app is best for this? But to be honest, I’ve been a little apped out of late. Sometimes, the tech can take the focus while the language takes a back seat, and that defeats the whole object. Too often I’ve spent time faffing with note settings and layout before getting down to the main event.

Flash Cards on Cue

As if on cue, our evening class Gaelic tutor recently prompted the group to dispense with the tech and go old school. Our homework task was simply to create paper crib notes for the material we were finding trickiest, and set them in prominent places around the home. She calls them ‘bingo cards‘, by the way, proving that everybody in the world does seem to have a different term for these linguistic comfort blankets.

So, out came the colouring pens. I’m a fiend for new stationery – a predilection I’ve noticed is shared by a lot of us bookish linguaphiles. I had a fresh pack of Staedtlers just begging to feel useful. I knew it – they weren’t just an impulse buy, after all.

The Magic in the Doing

As with all these things, the magic is in the doing, as much as the result. Investing a bit of time and creative energy into your resources doesn’t half help you cosy up to your language. I was pretty loved up in my index card creations and their technicolour irregular verb decorations on one side, and English prompts on the other:

Scottish Gaelic flash cards with irregular verb paradigms

Homemade Scottish Gaelic flash cards with irregular verb paradigms

I must admit, I didn’t overthink (or even plan) them. Rather than faff, I just had fun. The colours don’t have any special significance apart from separating tenses from each other. But it doesn’t matter – they say little things please little minds, but I was quite content to keep my mind little and my thinking nice and simple with them.

The verdict? They’ve already helped me in Gaelic convo starters – a lot.

Sometimes old school really is the best school – especially when it provides an excuse to buy more stationery.

Question marks after a mix-up. Image from freeimages.com.

Mix-Up Management – Laughing Off Word Confusion

We all get things wrong when we’re learning a foreign language. Given that we get things wrong in our first languages all the time (just look at Malapropisms and Spoonerisms), there’s absolutely not a drop of shame in that, of course. What’s more, the odd comedy word mix-up can be a rich source of effective educational moments. My go-to personal anecdote (retold so many times I won’t go into detail here) involves the German words Durcheinander (mess) and Durchfall (diarrhoea). Enough said.

Mix-up bloopers are at the fore of my mind lately, as they keep cropping up in Greek. Like German, Greek likes deriving native vocabulary with prefixes, so there is ample scope for someone like me to make repeated comedy slip-ups. The perfect funny word mix-up doesn’t always have to be with near-identical words, either. In fact, the most hilarious of them are usually only similar in a shared prefix or syllable. Here are some of my most frequent howlers of late:

κατάσταση (katástasi) – situation κατάστημα (katástima) – shop
σημαίνει (siméni) – it means συμβαίνει (simvéni) – it happens
απόγευμα (apóyefma) – afternoon αποτέλεσμα (apotélesma) – result
θεός (théos) – god θείος (thíos) – uncle

The greater the difference in meaning, the funnier mistakes can be. Depending on your appreciation of inappropriate humour, they can lighten the heaviest of moods. One of my best recent Greek openers is:

  • How is the coronavirus shop in Greece at the moment?

The Same But Different

But there’s another class of mix-up which has little to do with how similar the words are. Instead, the confusion occurs when the words have been learned together, creating a kind of messed-up context effect. A somewhat embarrassing recent example of mine involves the pair:

  • θαυμάσιο (thafmásio) – wonderful
  • απαίσιο (apésio) – awful

This antithetical pair isn’t particularly similar, save the -σιο (-sio) ending. The problem is, I learnt them together in the same chapter of Teach Yourself Greek about twenty years ago. And because of that, they’re united in unholy matrimony forever more.

You can see where this is headed, friends. They’re obviously both great words for reacting to another person recounting news or a personal story. During a recent Greek conversation class, my teacher was explaining how his elderly grandmother has quite serious diabetes, which is why she is shielding during the coronavirus crisis. My reaction?

Wonderful!

Oops. Good job he has a good sense of humour too. It wasn’t the first time, of course – that was responding to news of an earthquake. And for sure, it won’t be the last.

The Mix-Up Upside

Inappropriately funny outcomes are actually great news, though. No, honestly, they are! After you crawl out of the hole in the ground you fell into, these moments actually create huge extra salience in your memory. Weight is attached to them. You easily remember those laughs you had over comic misunderstandings. Sometimes you even give them names. With my Greek tutor, I’ve dubbed the shop-situation one ‘the Richard error‘, for instance. I will never forget those words now.

That said, if they prove persistent, it can be a good idea to be a bit more active in ironing them out. For me, the best remedy is simply ensuring to drill vocabulary in context rather than as isolated words. Learning the Greek for situation in a full sentence provides further sound cues like prosody and rhythm, which ultimately cements the word in memory along with its active usage.

One method for this is customising Anki vocab cards to include a field for an example sentence. For native-translated source sentences you can turn to a simple Google search, but I particularly like to use the Tatoeba corpus site for its language-specific search features. Corpus engines with mass sentence archives from subtitles, like Glosbe.com, are also extremely handy.

An Anki card in Greek with a sample sentence to help avoid a mix-up with other words.

An Anki card in Greek with a sample sentence to help avoid a mix-up with other words. This is another from the mass of kata- prefixed words that can be so tricky!

Talking of comedy, one of the most fun things about subtitle databases is their out-of-context hilarity. My sample sentence above, for example, means: “Babies are not good at building roads, look.” I haven’t a clue where this comes from (and to be honest, it’s probably best left a mystery). Needless to say, it makes me chuckle every time it pops up. After all, it ticks the be a clown box, and that is no bad thing when learning languages.

In short, if you find yourself confusing shops for situations, laugh it off – that humour is doing your vocab memory a favour.

 

Pidgins - or pigeons? Picture by Lozba Paul, freeimages.com

Feeding the Pidgins : Perfectly Imperfect Communication

One of the language learning lifelines that has kept me going during lockdown is our little Gaelic chat circle that meets weekly on Zoom. We started off as an in-person pub chat group back in January, but as normal life started to shut down in March, our ever-enthusiastic organiser decided to keep us going in cyberspace. Thank heavens for organised folk.

Our the months, the pendulum has swung back and forth with numbers, as is always the case with these things. Some weeks we manage a proper little group chat, and occasionally there are just two of us. But there is always someone there, and the determination never fades: nothing but Gaelic for half an hour!

Perfectly Imperfect

The remarkable thing is that none of us are remotely fluent. In fact, most of us are hovering around A1/2, with our main point of commonality being the Duolingo Gaelic course.

How on earth do we manage?

Not badly, all things considered. We communicate enthusiastically and fluidly amongst ourselves, gossiping on all kinds of topics from home life to politics. To do that, we do supplement, where we have to, with the odd English word or two. Feumaidh mi a dheanamh an washing up a-nis! (I have to do the washing up now!). A bheil lockdown ann a-rithist? (Is it lockdown again?) But we have a good online Gaelic dictionary loaded up in the background to share any pertinent new vocab.

We might sometimes use our own loan translations too, like “coimhead sgìth” (literally “looking tired” using the verb ‘look’ instead of something more idiomatic – probably with coltach!). Imagine our happy surprise then, when it turns out that some of these made-up-on-the-spot forms are attested and used in first language speech too (no doubt due to the influence of English, mind). I do sometimes shudder to think what a pedant or purist might think, listening in.

But still – it works!

Pidgin Fanciers

What we’re doing feels, in some ways, like the creation of a pidgin. Just like our peppered-with-English Gaelic, pidgins arise from the need to communicate using limited knowledge of a base language. Just like grander-scale pidgins, more than two languages can end up in the mix too – a couple of us have some Irish, so that gets thrown into the pot as well.

In essence, we use what we have to say what we want.

The upside? We have become really good at that handful of colloquial structures we all share. An toil leat…? (Do you like…?), an robh thu…? (were you…?), nach eil e…? (isn’t it…?) They are all pretty much ingrained now!

But I know what you’re thinking: but what about all the errors and mispronunciations being reinforced without any correction? As real-life pidgins progress, the divergences from “standard” grammar may crystallise into something new and more ordered: a creole. As creative as it sounds, that isn’t quite our goal.

Fortunately, we have a couple of safeguards.

Taming the Pidgins

Firstly, we do have a very competent speaker who attends quite often, who has been a brilliant source of guidance and advice. Secondly, a couple of us still attend formal Gaelic classes as well, so there is always an external guiding hand to keep us on the straight and narrow.

Finally – and anyone can do this, even without access to more knowledgeable speakers or learners – we note down anything we are unsure about during conversation and pledge to read up on it after the chat. Whether in textbooks or via a Google search, the info you need is never really find.

In short, don’t let a lack of vocabulary and grammar knowledge stop you from trying to speak a language. Have a go at feeding the Pidgins. As for us, we’ll certainly keep on throwing them crumbs – it’s got us through two lockdowns, and it’ll get it through the next!

Are you looking for some more Gaelic resources after exhausting Duolingo’s course? Check these out!

Building Blocks. Image by Jeff Prieb, FreeImages.com.

Building Blocks for Faster Fluency

The highlight of my language learning week was a short, spontaneous dialogue in Swahili. Before I get too big for my boots, I should add that it was about buying bananas, and wasn’t based on fact. Rather, it was invented on the spot in a university conversation class. But the point is, I coped with spontaneous conversation after just two or three weeks of learning a language. You can too – it’s all down to building blocks.

So what is a building blocks approach to language learning? It might be best to define it first by what it is not. Learning via building blocks is the opposite of rote phrase learning. Instead of static, clunky chunks, it focuses on mastering a limited but optimal set of words and phrases to combine in multiple permutations of useful sentences.

It’s not quite the same as learning an exhaustive grammar of a language, which is the longer-term route to manipulating language spontaneously, rather than relying on stock phrases. The difference is that building blocks learning focuses on efficiency, favouring the most useful bits and pieces to get you up and running super quickly.

Ready-Made Building Blocks

Unsurprisingly. whole language learning techniques have been built on the principle of shuffling basic blocks around. One of the most familiar from the bookshops is the Michel Thomas method. These use a chatty student-teacher format to gradually introduce simple building blocks, and invite the student to play around with the cumulative result. As such, the real skill students gain is the art of sentence creation on the fly, rather than plain old parroting. I’ve found them fantastic introductions that get students communicating in full, novel sentences extremely quickly.

Recently – big thanks once again to the lovely folk on the polyglot social media circuit – I found out about a whole bunch of free, enthusiast-authored courses that also follow this magic blocks system. The Language Transfer channel on YouTube hosts a whole set of language courses, from the author’s native Greek to – yes, you guessed it – Swahili. They take a model learner through a whole set of jigsaw pieces to spark immediate, spontaneous communicating.

Custom Blocks

So how did the building blocks approach play out in my Swahili class, and why was it so effective?

Swahili verbs lend themselves to a ‘slot machine’, or ‘lego’ type approach, as our tutor likes to put it. You can easily swap in and out a very regular set of morphemes for person and tense. Knowing just ni- (I), u- (you), a- (he/she), and -li-/-na-/-ta- (past, present and future tense markers), plus a handful of verb stems, a learner can express a huge amount in Swahili. This is the ‘permutation strategy’ that makes knowing just a little bit of language very productive. And every language has hooks like this.

The Swahili example shows building blocks at the tiny end of the scale, working with little bits of words. At the other end, larger chunks like ‘opinion blocks’ can be a great boost. In Greek, for example, I like to chat with my tutors about what’s going on in the world. A hefty topic, you might think. But in reality, it’s enough to have a stock ‘building set’ of a few phrases such as “I like …“, “I don’t agree…“, “… annoys me” and so on. Like those Swahili lego bricks, you can build whole conversations out of those spare parts.

Banana Split

The proof of the pudding – or the bananas, in my case – is in the eating. I’m really pleased at how much I managed to say in Swahili after a couple of weeks of this process. And it’s all down to those building blocks, and an effective teacher who makes great use of the technique.

If you’re about to start a new language, consider giving one of those courses a try. And if you’re struggling to improve your conversation in an existing skill, try chunking it up a bit into home-made building blocks. You will simply go bananas at your progress.

Intonation adds a thousand different colours to speech. Coloured glass. Image by Simon Jackson on FreeImages.com

Intonation Training: From Yam-Yam to Yia Sou

When you meet me, one of the first things you notice is probably my accent. Despite being embedded in Scottish life for over a decade, there’s still an unmistakeable Midlands lilt that persists. The vowels have flattened out to something a little more neutral over the years, it’s true. But it’s in my intonation that you can still hear the imprint of my roots.

Midlands accents get a bad rap. Full-on Brummie, for instance, still battles to be taken seriously after years of parodies and comedy sketches. And the baggage that people attach to your variety of speech can weigh you down. That pressure is one reason many of us subconsciously begin to change our distinctive sounds when we move away from our home regions.

One thing has proven extremely resistant, though – that characteristic rise and fall, up-and-down, sing-song intonation of my West Midlands English. In the Black Country, where I grew up, that particularly strong swinging tone has given us some national fame as yam-yams (most probably from the local form “ya’m” for “you are“). The almost musical nature of it is something it has in common with certain varieties of Welsh English.

But as endearing as it can be to us locals, it can play havoc with your foreign language learning.

Intonation and Learning Foreign Languages

The reason is the same phonological interplay that anchors our foreign language speech to our native phonology. Just as much as our vowel shapes and consonant articulation, intonation is highly ingrained in our oral muscle memory.

The unwelcome interference stuck out like a sore thumb in my recent learning on the mass sentence training platform Glossika, which I’ve been using to improve my fluency in a couple of language projects. The great thing about this platform is the chance to compare your own pronunciation with native speakers’ renditions. But be prepared: it can be very revealing. I realised that my intonation in Greek – especially in questions – was completely off.

What was going on?

Well, it all comes down to my deeply rooted Midlands twang. The tendency I carry over from my own native accent is to go up at the end of a sentence. That’s not just in questions, either. If you listen to Midlands English, you might well notice that our intonation rises at the end of nearly every sentence!

Not so with Greek. Often, the intonation will fall after rising towards the end of a yes-no question. It’s a bit more complex than that, of course, and there is much more detail in studies like this one if you need the nitty gritty. But generally, it is quite a bit different from English (especially mine).

Training It Out

The solution, of course, is more of the tool that shed light on the problem. Plenty of reps later on Glossika, and my question intonation is starting to improve considerably.

Repetition is the key, here. And if you don’t have access to Glossika, it’s not difficult to make your own DIY solution using the mass sentence technique. First of all, you need to source neatly chunked, model sentences in audio format. This can be surprisingly easy to come across. Many phrase books, for example, come with an accompanying CD or MP3 download links. Often, this material is available for download without even buying the book. Audio support for German publisher PONS’ mini courses, like this Croatian introductory text, is one such freely available resource. Multilingual sentence repository Tatoeba also includes many native recordings for its entries.

Once located, you can organise the material as a playlist in the player app of your choice. Having them loop round on a reel isn’t far off doing audio-only reps with a rep tool like Glossika. While it won’t quite follow the very effective, high-frequency high-representation corpus method of that site, it isn’t a bad substitute to give the technique a try in working on your intonation. There’s a plus side to phrase books, too; they tend to include lots of questions, which is ideal if you also struggle with that particular aspect.

Bit by bit, my up-and-downy Midlands intonation is disappearing from my Greek. It’s a lot less yam-yam, and a lot more yia sou. As for my English? I’m older and wiser enough now to stand up for my accent. I’ll carry that intonation with pride – as long as it leaves my other languages alone!

A picture of a mouth articulating. Accurate phonetics gets us close to sound native. Image from freeimages.com

Phonetics Mismatch – Why We Mispronounce Foreign Languages (And Why It Doesn’t Really Matter)

This week, I had the great news of an offer to study towards an MSc in Linguistics. And, keen on preparing well for a good start, I started working through a couple of the set texts. First up: phonetics and phonology.

As language learners, all of us have probably touched this strange world of symbols and tables. Many materials will use the International Phonetic Alphabet (IPA) or an adapted version of it to describe how to make the sounds of a foreign language. Some grammars for intermediate learners, like the Routledge Comprensive series, often include whole sections on the phonology (sound rules) of the language.

Phonologists use precise scientific methods to map out all of these sounds and their interactions, a process that can take years. The thing is, as learners, we rarely approach the ‘everyday phonology’ of a foreign language scientifically. Most beginners will not see /a/, and think open front unrounded vowel. Instead, we listen to a model, and make an approximation towards the sound through mimicry.

It really is just an approximation, though. And diving into some phonological descriptions of languages I know (like this technical summary of Norwegian), I got a sinking feeling. I realised just how approximate some of my pronunciation was. How was I getting it wrong for so long? And why did nobody tell me?

The lesson in this is how often even very proficient speakers of a foreign language articulate words differently from native speakers. But how do we get so close – close enough to operate fully and comfortably in the language – without quite hitting the mark?

Best-match phonetics

It is all to do with what articulation tricks are most readily available to us – chiefly the sounds we learnt as children in our first language. They give us a shortcut to make a comparable sound via a slightly different route. And they are so ingrained, that we are often swimming against muscle memory when we attempt to learn a brand new means of producing similar sounds.

The trouble is, our native languages often lack an exact equivalent in the target language, so we draft in the nearest match, often without conscious awareness of it. For example, speakers of English learning Spanish (and vice versa), are a case in point when it comes to the phonemes /d/ and /t/. Most varieties of English realise these as alveolar stops – that is, with the tongue touching the ridge just behind the teeth. In Spanish, on the other hand they are usually dental, with the tongue further forward, touching the teeth.

Try and make both of the alternative /t/ sounds yourself with the word tin. Difficult to tell, isn’t it? So much so, that we barely do tell the difference when we first encounter the foreign language. Instead, we produce the sound using our native inventory, substituting the sound from our first language knowledge without even realising it.

Subtle things like this, of course, are what give us our foreign accent when speaking other languages. And of course, although we can strive to minimise a non-native accent, it is nothing to be ashamed of (quite the opposite, in fact!).

Native issues

Take comfort from the fact that the same substitutions happen in our native languages too. For much of my childhood, I struggled to say the /θ/ in words like thimble and think. In the end, I decided that /f/ sounded close enough (and nobody seemed to mind at first). I stuck with that approximation a lot longer than my peers, until it was finally picked up by a teacher at primary school and normatively squeezed out of me.

However many extra years of give and take, though, the process of initial language acquisition is a fantastic feat of the mind. Children rapidly discover phonetics inventories and the phonological rules that can take academics years to map out – and foreign language learners years to assimilate.

With that in mind, cherish your approximations. They draw upon all the cumulated skills of those early miracle years of language acquisition. And even if the fit isn’t quite perfect, the act of repurposing them in second language learning is still a wonder of brain gymnastics.