Where does artificial intelligence fit within the realm of ESOL? Is the fear that AI could replace language teaching jobs unfounded? Where does AI need improvement? Join us in this episode as Brent and Ixchell discuss the current impact of AI in language teaching and share their predictions.
The DIESOL podcast
Digital Integration in English as a Second or Other Language,
Episode 59, Artificial Intelligence in ESOL
Welcome to DIESOL This is episode 59. We are your hosts. I am Brent Warner
Oops, I was gonna be Brent Warner.
I was gonna say “I’m Brent Warner” (laughter)
That’s how my passcodes have been getting out all over the place.
Alright, the mystery is solved. How are you Brent?
Growing my beard daily
Living alive, good!
Living a lie as Brent Warner,
Is it still cold for you?
Yeah, it’s still cold over here. But it didn’t snow. It hasn’t snowed recently. If it doesn’t snow, it means I can go out and run. And as you know, I run to the castle Komaki Castle that’s nearby. And running makes me happy. I get to see foxes and cats. And old people feeding cats.
You say “I see foxes and cats.” That sounds like a like a Miyazaki movie or something.
It is that’s how I feel. And it’s dark and creepy. And there’s like little, little, I don’t know if they’re called the little mini shrines that are everywhere. Little tiny ones. And it’s dark. And I go at night and the air is fresh and cold. And it hits your face. And you’re like, Ah, I’m alive. Very Zen.
It is! (laughter) How’s california?
I can connect with that a little bit. So I’ve been I’ve been going to campus two days a week to teach. I’m just teaching from campus just trying to not be at home all the time. And but I’ve been riding my bike there. So I’m good. That’s that’s yeah, getting a little bit of exercise. And it’s, uh, it takes about an hour to ride my bike to campus from here and so. But I did it yesterday. And it was super, super windy. No, it was foggy out. And so Oh, yeah. We live near that coast. Yeah. And so like going leaving at like 6:45 in the morning. And then like, I was just drenched. Like, there was the whole my whole jacket. It was just completely soaked. My beard was soaked my, like, my, my helmet has a little brim on it. And like water was dripping. It was so it wasn’t raining. But it was just like, yeah, yeah, so I bet
I can find a beard protector in Japan. Or not? The Japanese don’t tend to grow beards. But if they did, I bet you they’d have one available
Isn’t kind of the purpose of having a beard to be the protector of your face? Like in theory.
Yes. So your beard was effective.
Yeah, it served its purpose. It protected my neck. (laughter) This is nonsense. absolutely nonsense! All right. What do you got for us?
We’re talking about artificial intelligence. Artificial Intelligence.
Let’s jump over. What’s artificial intelligence?
Artificial Intelligence? Well, no doubt we’ve heard of it. We are familiar with facial recognition. Of course, that’s like the number one people usually guess. We if your phone opens up, when you show your face, then that’s facial recognition. We use it for payments. But it’s also used in self driving cars, which of course, that’s the other big thing happening. Doctors use it to predict certain things with big data. And then we have of course, Google Translate and chat bots, which now are everywhere. Whenever you go to a website and you’re trying to, you know, find answers to something there’s usually a little chat, chat bot. But artificial intelligence is a computer or a program trying to learn from its own data. I don’t know stop.
I’m gonna let you define this as if the listeners like don’t have any idea what the idea of artificial intelligence is like, I appreciate your, your effort here to like to the one person who has been living under the rock and suddenly found themselves subscribing to the DIESOL podcast (laughter. Like, hmm, what is this thing called artificial intelligence? (laughter)
We don’t really we don’t really need to talk about what it is.
No, what we do need to talk about is artificial intelligence in our field, right. So we’ve talked about, you know, I mean, we’ve mentioned things like You know Siri and Alexa and some of those like digital assistant type of things, right? So the those are listening to you and trying to understand and supposedly getting better at understanding you. I don’t always agree with that with the Alexa. You know, Alexa diocesan to me and I get mad at Siri doesn’t
get better. It gets better at knowing at predicting what I might ask, but doesn’t get better at understanding what I’m asking. Even, you know, sometimes I asked, What’s the weather? And she’ll ask me to unlock first like, I know. And then by that, yes. And then by then I have to ask the question again. So she doesn’t keep track of her like two step commands, or three step commands. She can’t do that yet.
That’s, that’s pretty seems pretty low level to me. I mean, I’m not a not a programmer. So maybe some would give me a hard time for saying that. But it seems like like, you shouldn’t want you shouldn’t have to unlock it. Just ask it. But I guess maybe someone could drain your battery if they started just talking to Siri with your phone locked. Maybe? If I were getting to yes. That’s interesting. I
didn’t think about that.
Anyways, I turned Siri off as soon as I get my phone so
serious for me, too. Yeah,
serious, terrible. All right. So what we really do want to talk about though, is this idea of like using artificial intelligence, how is it going to be used in our field? How is it currently being used? I think we’ve got a number of questions. And so today, Ixchell, I think our conversation is going to be a little bit different. Because most of the time we we bring up some articles, and then we kind of talk about the research on what’s going on. And then we talk and try and share some tools and some practical things in the classroom. But today, I think we’re kind of going in a little bit of a different tack, and we’re going to be focusing more on well, we’ll talk about some articles and some things that are out there. But I think what we’re going to do is then talk about maybe some predictions or what the future might look like, potentially, with some some of the artificial intelligence changes as they start coming into teaching and into Sol and all those other things.
So we’re gonna talk about, maybe talk about some questions that come up in our field. I think the number one question, and it seems to come, obviously, from a place of fear, I would say is, well, artificial intelligence replace language teachers?
Yeah, I think we’ll get into this later in their predictions, I think. But I’m gonna tell yes.
I am going to say don’t think we’ll fully replace them. I think they will be another part of there’ll be another tool? I don’t know. I can’t see that far ahead. All right.
Let’s let’s hold that question until the end, because I think there’s some conversation around that one. But some of the other ones we can kind of step for prepping in. So what else we do?
In our field? Where is artificial intelligence currently being used?
So one of those, this is a good point to get into is like, I think one of the places we see artificial intelligence being used right now is a lot in Google. Right. So we’ve got Google Translate. We’ve got the voice recognition software. So we’ve talked about this in the past and I talked about in all my presentations and things is like the Google Voice Tran transcriber, right when you’re in Google Docs, and use the voice thing, and it writes down what you’re saying. There’s also you know, a lot lots of these programs or apps that are out there that are like trying to help you with your language and kind of using algorithms and trying to respond to you and teach you vocabulary words, a lot of those things are using AI algorithms to get them going. I guess we could be a little bit careful. And we’ll we’ll try and be kind on ourselves. Because some true computer scientists might be like, well, there’s a difference between artificial intelligence and machine learning and you know, like, as machines are actually teaching themselves versus just recognizes what is programmed to know what kind of say, Okay, we get that kind of, but, but I think for now, we’re just kind of trying to think of like the computers responding to us, right, and that is interacting with us in certain ways, without another person being involved with that. So those are a few shell any other ones that you’re thinking of? Yeah, so
I know that we have a few extensions that help students to paraphrase. And that, you know, obviously, there is some kind of learning going on if something is able to pick up what you’re saying and then giving you the different options based on that. So I think there’s a extensions that do that I don’t I haven’t tested extensively, because you know, I use it mostly for academic writing and short snippets of text, rather than longer snippets of text. And but there’s plenty of websites that I think one of them was called resuIt. Hammer. Yeah, we’ve talked about Yeah, I remember you talking about that one. And that one, just kind of, I’ve never used it, but I remembered it. But stuff like that, that addresses one component of language. But then imagine if all of them came together, right. They will translate those transcriber extensions all in
one. Yeah. Yeah. And I think I think that’s where the creativity might start coming, as we’re starting to see, like, how do we blend these different ideas together? And so I think that there’s some, some fun stuff to play with. And then we’ll also get into like, you know, what are the implications for for our field? Right? What are the implications overall? I think kind of what you’re talking about, right? There’s going to be career correction stuff, there’s going to be things like, you know, interactive programming, there’s going to be fun, cool. But let’s talk about a couple of these articles here. So I’m not sure if there’s any ones that were stood out to you that you want to talk about, I kind of had one or two points from some of them that I thought were interesting.
Yeah, well, first, I wanted to point out that as I was searching for information, a lot of the stuff that I kept coming up with was, you know, material, pointing at some kind of aspect of AI, fused into some kind of software, and then that research or the studies or whatever it was a data gathered was by that company who was. So it was really hard to find something that was
amazing. what our program does is the best.
Yeah, so I found a nose that that tells me that there’s not enough out there. And it’s quite obviously quite limited. And only people who test that specific thing can give you feedback. So I you know, I had a hard time with this. But you found a couple of articles. Yeah. So
there were a couple out there. One from well, actually a couple on this are this is what you’re speaking about. There’s this company called Glossika. I guess, and I haven’t, I haven’t played with this yet. But but you know, the way they kind of describe it in one of the articles is they say like, and this is a quote from an article it says unlike traditional approaches, which utilize lessons and weekly plans, Classica attempts to tailor learning material to your specific level and needs, users select the topics that are most important for them and work in a comfortable pace on their desktops or smartphones. And so that kind of seems like a smart way. You know, it’s kind of it’s also similar maybe to Duolingo. But I think one of the arguments that they’re making in here is that a lot of these language learning things will start you off with language that you don’t need it for your practical work life, for example, like a lot of people are studying a language because hey, I have to do some business, right? I have to go fly overseas, I have to go to this conference, I have to communicate with people. In that case, you know, do I need to know how to say, you know, postage stamp or, you know, like, giraffe or you know what, like, there’s some of these things that are not really going to be relevant. And so I think they’re trying to get into maybe a little bit more and then and then try to get you use the AI to customize towards what your actual outcome learning goals are, which seems like a pretty smart idea, right? Like, this is what we do in a lot of like, specialized English classes, right? It’s like the students are like, Well, I kind of have a little bit of a base, I don’t really want to just learn all general English anymore. Now I need to kind of get it specified towards my meats.
Right. And I think that, you know, points me to the fact a problem I keep running into here in Japan, where I have studied Japanese enough basic Japanese to do superficial daily kind of things. But then I missing the piece that allows me to form a deeper connection with someone that gives that shows my personality. And I keep finding apps that will help me with Japanese, but they’re mostly helping me with basic vocabulary sets or very basic Japanese structure, which is what I already have, or and then it jumps into, like people who want to be who are living here who need to be fluent, like fluent in all the areas and so there is a gap between the beginner and and someone who might be more proficient. So I ended up having to reach out to an actual human tutor and someone who was giving classes and I just gave him a rundown of like, here’s what I’m looking for, and that’s what I would want and something that you know and AI offers that, as opposed to like Duolingo, where, you know, their, their approach is teaching you certain things as they would to a child, you know, when they’re first learning and acquiring language, their only native language, or the vocabulary. But something where it’s like, hey, I need English for the workplace. And a little bit of, you know, I need to be able to build relationships with people. Yeah, that’s the kind of specialized component that I think AI might be able to do.
Yeah, I agree. So another one that I thought was interesting was this Forbes article? Ixchell, did you look at this one, the language lessons from artificial intelligence that came out last year? In and go ahead, yeah, this one’s turned to talking about this company called our one and pre show, I showed you a little video of this. This is pretty amazing. So so it got into this whole thing where like, basically what they’re doing is they’re scanning an actor’s face or an actresses face, right. And they’re, and then they’re, they have all these different voices. And so what they do is they put in a script, and then this voice, actual, this voice and face can be made into a video in order to talk directly to people with just like the content that already exists, right? And so you might start by going well hold on a second, if you’re going to type out all this material, then why why wouldn’t you just record it for yourself? So the idea behind this, and if we’ll put the link in the show notes, it’s really worth going and taking a look at that. It’s pretty cool it up. It’s kind of creepy. And it’s kind of cool. But like, Yeah, but but basically what would happen is, let’s say that someone liked me as their teacher, right? They’re like, I want Brent as my teacher. And I don’t have time to be everyone’s teacher, right. So what they might do is they might scan my face, and they might kind of put scan my voice in there. And then they put in all of this general material, but I’m the person or which, you know, this artificial intelligence version of me is out there talking and explaining all of these things, when all of these scripts maybe could be used for different purposes. But the script, the language itself, could just be collected, and then put together right in there. And so you could actually have this interaction with a person that you like, and sorry, I guess I’m making the assumption that people like me, sorry. But you could kind of do that, right. And so again, this is step one, where you’re really kind of talking low level stuff. But if you start getting into this, I actually really see a future where there’s gonna be some teachers taking advantage of this, and then having a huge database, because you basically start copying and pasting text right into this thing. And it just you click a button, and it’s like, boom, here you are, right. And here’s, here’s you, or here’s this artificial intelligence thing, going through and talking through the script, but also catching intonation, hopefully, I’m not totally sure how great it is, and all of those things. But as it will continue to improve, then you’re going to have this fully customizable thing. And Ixchell, we were talking about pre show with the remember the GPS is
you can change the voice of your GPS to like Buzz Lightyear from Yeah, from Toy Story. Yeah. So James Bond,
exactly. And so can you imagine like, people are just going to go, Okay, I want this celebrity to be my English teacher now. And I’m going to take a lesson and they’re going to kind of just talk to me. And it might not be totally interactive. At this point. I do think down the line, I think there’ll be like a real ability for response and things. But it’s pretty interesting.
You know, as you’re speaking about this, and I don’t know why the thought that it crossed my mind. But you’re talking about collecting this speaking samples from a teacher and their, whatever methods they use to teach or explain something that’s already written and you’re just putting it in there. This is no different than music programmers who use different tones and someone’s voice, collect them, and then play them and make up an entire song with a 3d avatar. And then you’ve got Hatsune Miku, who has like a following of millions of people all over the world, and she’s not real.
I was thinking of the program and all the the music talking. No, and I mean, but yeah, I
mean, if we’re talking about English teaching, you’ve got to be able to hear the person’s intonation. And yeah, it’s been done with music. So then not but now you’ve got a face that exists and you can have Brent teach you recipes.
For the low price of 999. I’ll teach you a recipe. What recipe?
Read me a bedtime?
I don’t have to prepare because, yeah, you know, it’s funny. I’m like, Oh my gosh, I have to prepare a recipe. The whole point is that you’re not preparing anything. Yeah. Okay, so pretty, pretty amazing. Pretty cool. Yeah, but definitely, I hope that people go and check out the show notes and get the link to that, because it’s worth watching just the intro video, and kind of getting a sense of where that that might go.
So I picked out a quote from one of the articles. It’s a Pickhardt 2020 article called intelligent information processing for language education, the use of artificial intelligence in language learning apps. And of course, we were talking about Duolingo. And you know, recently Duolingo became an official English tester, or, you know, like, the equivalent of a TOEFL really, and you get a certificate. Yeah, you get a certificate? Yeah, I think it’s been now two years or so. No, that’s
not very recently, two years, but that’s COVID time. So like, so like, 20 seconds, right? That’s what you really mean. Mm hmm.
But anyway, um, so here’s a quote, because we talked about where we were finding AI, and it tends to be in language learning apps. The changes in pedagogy are underway regarding new approaches are both the educators who are responsible also for the creation of the courses, but also for the students who have radically different needs and approaches to retrieving knowledge and information. And this is the line that that that stood out to me, the current young generation of the students who were born around the year 2000. So Daniels, millennials, because Millennials have radically different tools to acquire competencies and skills, they also use different tools for evaluation of the information they have acquired online, these aspects must be taken into consideration, as as well, so as not to lose competitiveness and global sustainability. So it’s, it’s a warning, it’s the warning that we’ve always, you know, in us in edtech, we tell people, Hey, this is coming we need to get with the times, the longer you leave the gap, you know, the more it widens, and then you’ve got a pandemic, sort of shoving you through it. And so I think, to not talk about AI, or artificial intelligence is, we’re kind of maybe we’re not so sure about it. We’re sometimes fearful of it, but it is there with our, you know, self driving cars and everything else. It’s there. So I just thought it was important to, to highlight that this is why we’re talking about
Yeah, yeah. And so, as we kind of wrap up some of these articles and research, I do want to point out this, if you start looking up like English language learning and artificial intelligence, like the name is going to show up over and over and over again, which is Marina Dodie Dodie Govich, which is, she’s a Croatian and one of my one of my bloodline. It’s not my bloodline, but my people from from generations past somewhere. So anyways, she has written lots of articles from a long time ago. So like, you know, way before, like the current conversation, I think the article that I’m, I’m pulling right here is from 2007, or eight, or maybe, I don’t know, I can’t remember but, but she, she’s been doing all sorts of different articles on different parts of using AI for learning. And she would analyze tools that other companies had made and see how it works with the students. But one of the things, the quote that I thought was interesting from this one called speech recognition technology in language testing, does taking the test in an English speaking environment matter. Which, you know, by itself, the whole article was interesting, but just getting into part of this, this artificial intelligence part, I thought it was interesting that she said, an important pedagogical question is also when to correct an error. Given the nature of a question, it seems very important to decide whether an erroneous utterance contains a genuine error, right? And so I think one of the things we’re going to be dealing with is like, hey, students say that they want to get corrected, but as teachers, we know that we cannot correct every mistake, right? And so I think we’ll definitely see some major missteps, with some companies trying to correct every time a student makes an error. So let’s just say let’s imagine this kind of beautiful future where it’s like, I just talked to the computer and the computer talks back to me, but the computer will also stop me if it’s if it recognizes that I’m making an error, right? And so if I say a You know, something like, yesterday I, yesterday I go to the bank, right? And it’s like, okay, that is definitely an error and it’s wrong, and it should be fixed. But if that’s a, you know, if that is a one time error from that person, and they’re just kind of moving through a sentence trying to get more important information across, then is, are they going to get frustrated when the computer stops them and says, Hold on, you said go and it should have been went right or something like that. And so we’re gonna definitely see because even as humans, we deal with this right? We’re like, should I be fixing this? Should I be overcorrecting? Where’s where’s the right balance that actually helps them and makes them feel motivated to learn, but doesn’t dissuade them from their own abilities. Right. And I see some interesting, we are going to see a lot of kind of dealing with that as an issue for trying to have interactive language learning I think and and error correction inside of AI. So
imagine that so I’m thinking so how if there were this magical tool already there, how it would be correcting us as quote unquote, native
Oh, my God, terrible transcripts. So we get these AI transcript for the show, and I have to spend hours going back. I don’t get
everything does not recognize my name after how many times 59
shows it’s definitely not learning.
I mean, come on. Yeah. What’s always wants to change me to what, Michelle? Which is what people always wanted to call me.
Yeah, for sure. You are you are Michelle in there, or sometimes you are “he shall.”
No, “He shall” like “he shall go to the royal throne,” I guess
All right. I am gender fluid. And that is totally okay. Yeah.
But yeah, you’re right, though, like this is it’s, it’s very interesting to kind of see this and like, go, Okay, well hold on a second. Where is it going to? Where is this going to land?
Well, here’s what would be cool. Because I tend to run into this issue with some of my students who have studied, you know, they know every idiom in the book. But unfortunately, the idioms are old now. And yes, I know the idiom, I’ve heard it, but only older people use it. So it doesn’t seem to fit their students personality. So now this tool would have to take the current data. In the last I don’t know, I would say, like 10 years, because I don’t know how how fast language is changing, especially with social media. So you would want to take Yeah, you’d want to take like a period of 10 years analyze all that language, you speak something and now it tells you Okay, are you speaking casually? Are you speaking formally? Is it? Do you want it in the current slang version? Or you want? Do you want it in slang from the Midwest? Or do you want British? Whatever? Because that would be cool.
So it’s like, register and putting all of that into it and giving your responses and those way? Yeah, for sure. And that’s actually super useful, because like, let’s say, for example, and I think there are some things that do this, like it, let’s say you’re writing an email, and like, the tone of your email is not appropriate for like writing to your teacher or for you know, or like trying to get a job or something. And so, back when I lived in Japan, this, this colleague, who you know, whose job it was to, to send us messages and kind of, you know, keep us on top of our work, and hey, send us this information. And she would always use Command forms. And she’d always be like, yeah, you will send the paper to me by tomorrow. And
my resale? Yeah, I recently had this, someone come to me to double check their translation of a formal ceremony schedule. And the first event or the first item on the schedule was everybody rise, for, you know, to greet this, you know, VIP person, but the English version was all men and women will stand up. Yeah. And it just sounded like, Alrighty, then. So, and then it would shift to so but again, you’re translating in, you know, honorific language, you know, from Spanish, Spanish. I mean, sorry, I’m getting all my languages confused. But English doesn’t have an honorific form, right. I mean, it has formal language for a ceremony, but you don’t necessarily have all of the different you know, all the different levels that Japanese or another language might and so you would have to know that. Yeah.
Yeah. And then and then it’s like, okay, so how does it How does it recognize? Or how are we supposed to translate that into a way that still maintains politeness and the tone of it, all those types of things. So it’s,
I mean, that would be useful for us too, because as teachers, because sometimes we, you know, I catch myself, and I’m more careful with this now, where I don’t want to generalize. That’s how people speak in this way, because students will ask me, how do they say this in such state? And I’ll say, I don’t know. I’ve heard this. And when I hear it, I tend to think of it’s from the south. So we would have to check. But I don’t want to generalize like all people from the South speak this way or all because I may have moved around.
So like, yeah, kiss my grits.
You guys, you guys say y’all? Oh, yes. Only in Texas. But the truth is, I used to say y’all in California, and
all is showing up big time in California. So the conversation we were at a restaurant the other day, and like the two different one waiter when waitress came up and just straight up used it very naturally. Everything else. So it’s it’s y’all is making a a an invasion all the way across
Texans moving out of Texas. Is it flipping the script?
I’m serious. Yeah. And I think that’s also interesting, because it might be also part of the, the gender and the pronoun conversations with
Absolutely. Oh, yeah. So again, that’s a whole other you know, and when I talk about these movements, but these changes in language and how we’re adjusting to what reflects our society here, when I talk about it with like, say, my Japanese teachers, they’re not there yet in terms of social movements, and they’re not there yet. So for them to hear that, yeah, we now can add our pronouns. And it’s important to think about how you’re going to what, what you’re going to choose for yourself, they are not there yet. So for them, it’s a brand new thing. But it does help them to stop and think like, this is where we might be going in a few years as well.
There’s a lot of interesting stuff, but let’s take a quick little break. Alright, so is February, the month of love and all sorts of other good things. And so, for us, I think this time around, we’re just asking a favor to spread the love a little bit. And the love right now or
who I am, it’s not a Neil Young song.
Oh, are to gold. Alright, so if you, if you’ve been enjoying the show, like, I think what we’re gonna ask this time is just share an episode with a teacher that, you know, if you can share it out and say like, Hey, this is cool or useful or valuable. Whatever episode you like, or if it’s this one, and you’re on your pod catcher, and you can tweet it out or whatever. It would mean a lot to us to kind of share out a little bit say, hey, this show is useful, valuable. And that is our, our Valentine’s ask is can you or Valentine’s Day ask? We love to ask for Valentine’s quick thing like things specifically? I don’t know if thats –
DIESOL crew be our Valentine.
Oh yeah! We could Oh, I should we should have come up with some clever like, DIESOL puns around Valentine’s. We’ll work on it. Well, we’ll put up some Valentine’s Day cards and see what happens. If you’re willing, if you like the show, if you can share it out? Just one person would be awesome. Thanks so much.
Yeah. All right, let’s talk about predictions. I think this part is fun. Because there’s no limit. Buddy, you got Brent.
So I guess we’re trying to think about like different ways that things that might happen with the future of artificial intelligence and with education and, and, of course, our field for language learners. So I’m going to go with a lot more personalized content. And I think we’re starting to see some of this coming in already. Even with like, Duolingo and the the, the service that we talked about before, where we might be able to see things where it’s going to start sending learners information based on their direct interest, and also their language level, right? And so we’re going to be able to see things where it’s like, oh, you know, if you like, you like basketball, and you’re kind of intermediate to then they might start just sending you you know, an article that you can read like a real legitimate article, I think this is this is some of the things we’re gonna start seeing is like, they’re gonna go out there and find an actual real article that you could handle. And then it’s going to send it to you and say, Hey, this might be of interest to you, right? And all the AI would be doing would be analyzing the language of the article. So maybe Be at something like USA Today, where it’s like, a little bit easier to read, right? So it’s not going to be the economist, it’s not going to be the new New York Times or anything like that. But maybe it’s USA Today. And since it knows that you like basketball, it’s going to send you an article on Steph Curry or something like that. And I could see, like, that’s just one level of it. But I can really see people getting like write in their inbox or write on their apps or whatever it is just super personalized content, things that you know, that you’re going to like, because the algorithms are getting, you know, sometimes we complain about it. But like, I think the times we don’t complain about those algorithms that are chasing us, or the times when they’re actually being successful, which is more time. And so we’re like, what is this garbage our algorithm? It doesn’t know anything, but really, the most of the time, we’re like, oh, yeah, I’m actually kind of interested in this thing. Right. And so I think we’ll start to see, as it comes into, like language learning fields, that maybe maybe they can start to customize that material and send us authentic materials that will be right to our interest.
And right that was that was solved the issue of having to find an article on true crime, because I’ve got students who love true crime, and then an article on comedy or, you know, that type of humor, in the writing style that would really cut down on having to, you know, gather all that material, because you’ve got to, like you said, We’ve got to personalize the content. And if it’s, you know, if it’s English content, and the student is already showing an interest for that topic or that genre, then you would have you know, you’d have a cash at your fingertips of all of that.
Yeah, well, and also, I want to point out, I mean, I’m saying articles, but that also means podcasts, and yeah, film shows, all everything, any type of media content and digital content that these things can go out there and scan they can spend, send their bots in their web crawlers all across and just grab that information. web crawlers, like
crawlers. So So yeah, so it could bring that all back. Right. And so I’m, like I’m saying articles, but really anything? And could you imagine how robust your your content and your database of things that you just might be?
Yeah, and that’s, you know, it takes so much time for a teacher to find that information, and then find it at the level. You know, I’ve got a student who likes animals. So I found the documentary, but now I really want an article about, you know, that layers on top of that, and I also want a podcast because I want him listening. And then I want conversation questions. And so I’ve got to pull all of this stuff, it takes time. But that’s just one student. So I think I think my prediction is, and I know, we talked about this, in the medical field, there’s already you know, big data is being pulled to analyze potential medical treatments, through AI. And I feel that programs like Google Translate, or Elsa, they already have data that they’ve collected, they’ve got, especially if you’re translating with using voice function, which I’ve actually used here in Japan, and it’s not bad, it’s actually pretty good. But more availability of a program. That includes more more than just the translation part. Just expansion of what we already have in terms of tools. So I just feel that it ultimately, it’s going to be Google transom Google product that is going to purchase other things like Elsa, for example. I know, we’ve talked about LSI, before and integrated in there. And I think it’s going to be something that’s free, but we’re going to be trading off the fact that in order for this to have the data, it’s got to gather it from human sources, that that’s gonna be us. And that’s going to be you know, this whole privacy thing or, but I mean, I, I’m glad that Google Translate is at the stage it is now because 10 years ago, it wasn’t and it’s thanks to all of the people who have allowed to have their voice recorded or their you know, their key input. Catalog. Yeah,
yeah, for sure. So I think another one here, kind of, you know, in connection building on top of all of these things, I think there’s gonna be automated i plus one content. So like, if you’re gonna get that, you know, that comprehensible input, and you’re going to be starting to, to add, like just a little bit harder than what you’re used to, or a little bit more advanced than what you’re comfortable with. Right? And so I think that it doesn’t, so again, let’s say we go back to like what I was talking about with personalized content, are you kind of saying like, Hey, we’re going to be pulling in this big data is that then it it will start actually summarizing or paraphrasing or expanding based on a Little bit more than what it knows that you already know. Right? And so this is baseball. Yeah, so it’s like tiered readers. So like we’ve got new seller, right and new seller, they have people that are going in there and doing all of this work, I think AI is going to do this automatically. And it’s going to be based on anything that you are interested in as well, right? So it’s gonna be, it doesn’t matter if it’s something that oh Newsela has already gone in and figure this out. It’s like, No, I want to go into, I want to go right into Reddit, and I want to read this Reddit thread that I’m looking at, and I want to push a button. And it’s gonna change the language in there and make it just a little harder for me. So like, I can read it, I can understand it. But it’s also going to give me a couple of challenges right there, like built into the browser, for example. And let me kind of interact with it and say, like, I do understand what this is saying, or I don’t understand what this is saying. And so it did take, let you know, if we’re going with the Reddit example, hundreds of people’s posts, and in a matter of a split second, it’s gonna readjust that language. So it’s still saying the same thing, but slightly adjusted to be able to be a little bit, one maybe easier for you to understand, but to not so hard and not so slimy, that you can’t follow along with what it’s getting at. So I think that we’re gonna see some some area, I don’t know if it’ll look exactly like that. But there’s definitely the potential for that.
Can this potential tool, examine memes and then sure, translate them into what they’re actually saying? Because then my mom would understand that?
Well, I mean, you know, ultimately, like, that’s, we’re talking about is being able to do anything that he human brain can do. Right. So it’s like, yeah, so but it’s, that’s an interesting one, because I have listened to podcasts in the past where they, you know, like, on, on reply all back in the day, they used to do, like, what does this mean, mean, right? And then they would have to get into, like, full context and like, explain it for 30 minutes, which, you know, that’ll be a much higher level, but I could potentially see, like, here’s the background on this, and here’s the things you need to know.
I mean, ultimately, its etymology kind of, you know, because that’s what we do with words. Now we find the etymology and we see how they’re related. And we find it but it’s just etymology of, you know, 21st century language, I
guess, Edom emoji
pneumology. That’s it. We need to, we need to take that word now.
I’m gonna I’m gonna NFT that word and make some real NFDA. Wow, okay. Yes. That’s ours, except one out
there, we’ll buy it. I bet there’s someone out there that will buy it. So with all of this, all of this progress, I think that there will be more companies using this AI component as the selling point for their programs, which I always think you know, you’ve got to be really careful when companies are trying to get you to use their software or their tool, whatever it is, and using the one thing, you know, to as their selling point and trying to get you to pay more or painter’s description. And the reality is that a lot of this stuff may be in earlier stages of development. And I can see where this would be a big sell in the test prep world, because test, PrEP is not going away. And you know, everyone wants to pass the TOEFL and the TOEIC, and Isles and all those other tests. And not everybody can go to tutors or cram schools or etc. So I think that that might be a place where we’re gonna see it maybe spread more widely. But I also feel that we should probably proceed with caution because AI at this point, and with programs that pitch it as their their selling point, I don’t, it’s not a fix all. And we’ve got to just be cautious about it, I suppose.
Oh, well, 100%. This, this is we’re going to see this ridiculous waste of energy because here’s what’s going to happen. Students are going to get click a button to get AI to produce an essay for them. And then teachers are going to press a button to get AI to respond to the essay. And then students are going to click a button to get the responses updated and responded to and then the teachers are going to click a button to get recognize those changes in the response and get a grade and nobody will have done anything and nobody will have learned anything. I am kind of not joking here. So like let’s let’s keep an eye out for that. I mean, it’s kind of like dark dystopian, but like, but I really do see that as like another thing that we’ll have to be careful of you If we don’t actually have the human eyes on things and like, Where’s where’s the the human element to this?
Right? And moving on to creepier
so creepy or creepier. So we can kind of talk about this the deep faking world audio and video for faces, right? So we mentioned that before, like, I could scan my face and you could have me talking. I don’t know if you ever saw the seashell, but there was Jordan, Peele was like at an Adobe event of some sort. And they basically brought him on stage, and they’re interviewing him and talking to him. And then they bring out this other guy with a computer and he starts typing stuff. And it’s like, he like Jordan peels voice saying things that he has never said before, right? And then like, you see Jordan peels. This is a couple years ago, but he’s like, it’s like, what the heck is totally freaked out by it. And he’s like, this is, this is like, sounds super real. Like it caught all the intonation. So you could chop out like individual words, or add in words right in the middle of what he was talking about. So we definitely will see some room for deep faking. But then, like interaction with those deep fakes, so like, Hey, I don’t have a friend and I don’t don’t isolate that language, either. We’re like, we’re like, I don’t have a friend.
I want to get you back for posting all the time that I was puking during the show. Oh, that was
well, I mean, we were both anyway. So. So I mean, this is straight up Ray Bradbury, Fahrenheit. 451. Like I’m talking about, like, you’re interacting with a wall, right? You put something up you want to talk to it’s gonna respond to you. But it’s maybe it’s someone that you want to be your friend or you imagine, I think people people might actually go straight up crazy with this like,
BTS. Yeah, I want to teach me a language.
Oh my god, you’re gonna imagine that, you know, BTS, then you’re gonna go become a crazy stalker. But there is a real there, you know, like, I don’t want to be a Chicken Little. But there are also other other problematic areas, potentially. Yeah. All right, well, so what else do we got?
I just think this is a whole new version of what we, we I don’t even think we can fully imagine how language teaching and language learning is going to be with something like this, I think when we fear that our jobs are going to be, you know, destroyed by AI, or taken over by I actually think that it will just be a new version. And of course, we always fear that we’re not going to be able to do what, what keeps us what gives us a paycheck. But I just think it’s going to be a new form. And the same way that people feared comic books and video games and television, and the radio and texting and cell phones and fingerprint scanners and iris scanners and all that. I mean, CNN just had a segment on a flying car already being approved. And coming to the commercial market. Next year, next year. So
I am a little bit more on the other side of this. If you’ve heard me say this, I don’t say this to people, because they get mad at me about it. But yeah, I think it’s the end of ESL, I think, as a as a field, as you know, like for people that need to study it like academic ESL, and those types of things, I think we’re coming to the end, people who want to study languages for their own joy and for their own benefit, those people will still exist, there will be places to, you know, language learning will still happen. But I think that a lot of this stuff is going to get eliminated and reduced. And so I don’t think it’s going to be as big however, I do have one thing that is kind of a little bit of a light at the end of the tunnel, which is even, you know, some people really want to learn and understand the value of truly learning a language because the immediate response back and forth is something different, right? Even if you have like the best translator in the world, and you’re working at, you know, the UN and that person standing right next to you, and they’re doing live translation, that pause and that gap is a distraction to your communication, right? While you’re waiting for that person to translate and translate back and all those types of things. So I do see a place where people are gonna go, it doesn’t matter because even if I’m speaking and something is translating directly through a microchip in my throat and it’s actually speaking out as I’m speaking, it’s still going to be paused and not quite as fast. Because it has because the way that languages themselves work, right like the, the structure of the the grammatical structure of English has to be processed into, into the new language and it can’t come out at the exact second of the word that I’m saying because maybe their subject goes at the end of the sentence or maybe, you know, whatever it is and so like my punchline couldn’t come out or even a trance translation of a joke. highest levels right? That couldn’t come out immediately so I do think ESL as a field as we know it is going to die I know people are gonna go what but I really I Not now not today tomorrow but for the rest of your
life I mean but English English as a global language won’t die because I mean even if you could translate and use another trance you know a beautiful translation tool or device as you might see in sci fi yes films there are still ways that you cannot express yourself in a language because of the history of the language so for example why hasn’t Chinese Mandarin become the global language even though we know that’s where we need to be going etc? And that’s because you know, I was listening to this of course a podcast on NPR about how the ideas of freedom and freedom of speech being so ingrained in the English language you can express freely there’s again not that hierarchy of politeness according to who you’re talking to. So if I’m talking to the president, I’m going to say I’m sorry, if I’m talking to my dog that I just stepped on I’m gonna say I’m sorry if I’m talking to so so in other languages you don’t have that and so it might limit and it might be associated with certain aspects of communism or you know, our system ma cracy and so
speakers where there’s like language of God tied directly to spoken for sure there’s
Spanish Yeah, so English itself is not going away.
No, sorry. Yeah, I’m not saying English is going away I’m not saying English is a global language. I’m saying English as a second language as a field.
What I do hope this software or AI component or whatever it wherever it ends up being playing a role is that it also helps us monolingual signed bilingual but people who are monolinguals learn another language
because more time for it to right like to experience the joy of learning another language because it is a true joy right like understanding and coming and being able to communicate in those ways and just doing it for yourself. I totally agree with that for sure. So so I’m not going to let let my let the story end on a negative funeral no true joy
joy of joy Swift language learning joy all right, it is time for our fun finds. And this time I have something that has been helping me I’m a runner, a serious runner, which is probably why I injured my knee not stretching properly or warming down properly. And it locked up my knees and one of the ways that one thing that has helped me to recover my muscle healing is an EMS sheet and that’s electrical muscle stimulation. You do find them on Amazon but you find them for people’s abs like to define their abs but this is just the sheet you put on the ground. You spray it with water that’s that conducts electricity you put your feet on it and then you sit different frequencies and it actually exercises your it stimulates the muscles that may not be healing and of course it’s an item I found in Japan I didn’t find an EMS sheet on the American Amazon so I’m not sure if we have something I’m sure it exists somewhere by a different company but this one is by ATEX and it’s only about what $48 Which is pretty affordable and if you know for if you’re you know sitting down all day that helps you know helps stimulate muscles and yeah, so EMS sheet
Nice. Okay, so mine is a video that showed up online and it just made me so happy Have you seen this the slipper you guys know Okay, we’re gonna watch it together right now so it is so good so people who are listening this is purely visual thing
it goes on and on and on and he’s like he’s like trying to balance himself and he’s slipping and sliding and he’s like gets into like a running mode for a second and it is so joyous it is that
a tick tock video
well it’s showing me yes, it’s clearly a tick tock definitely put their their push out but yeah, it’s a it is a great video. We’ll put it up on the shownotes if you just I just keep watching it over and over again and just like it’s just putting me into such a good mood because it’s silly the guys that doesn’t get her like he’s just slipping and sliding all over this ice like a tiny little ice patch right in front of him and it’s like a it’s like the old like Laurel and Hardy like slipping on a banana like type of vaudeville comedy, but it is it is just so funny. So, so slippery ice guy. Yeah, that’s my fun find.
Well, thank you so much for listening to the show, we do ask that you share the show this month. If you’d leave us a review on Apple podcasts, they do help spread the word. And if you’re giving us a shout out any other way, just tag us we are on all the platforms on social media.
Oh, of course you can support us through our patreon or with buy me a coffee. We are grateful for anyone who is willing to do that. But like we said for February we are just asking share the show that’d be a great gift for us. So if you want to find the show notes, you can find them for this show and for other episodes at DIESOL.org slash five nine, the number 59 And of course you can listen to us at voice Ed Canada. You can find us on Twitter the show is at DIESOLpod and I should also point out we’re also on Instagram I never say that but we are we do we do post on Instagram so if you just want to kind of know that one new episode comes up. You can put it also at DIESOLpod on Instagram and I am @BrentGWarner
and I’m Ixchell at IXY underscore pixy that’s @ixy_pixy
In Finnish thank you is Kiitos. Kiitos for tuning in to the DIESOL podcast.
Thanks everybody. Take care. Happy Valentine’s Day
- Artificial Intelligence in English language learning (2020)
- Is Artificial Intelligence the Future of Language Learning? Glossika and the rise of AI-guided language study (2017)
- Language Lessons From Artificial Intelligence (2021)
- Intelligent information processing for language education: The use of artificial intelligence in language learning apps (2020)