has anyone used duolingo for an extensive period of time and can confirm it helps with actually speaking a new language? my experience with tools like these is that they help build vocabulary but you can't actually hold a conversation for a long time...
One of the hardest parts of learning a language is just that building enough vocabulary so you can start stumbling your way through speaking it. Duolingo, Drops and similar are never going to be "the one tool" that you need. But they are a useful way of building some of that initial knowledge to then start reading (childrens books), having horrifically slow bad conversations, and the other things that allow you to learn the language.
As someone who followed the same method (DuoLingo for vocab, start reading, start talking) and now speaks a second language fluently, I'm not OP but I'll go ahead and vouch for their comment.
Completed the tree, but admittedly it was shorter back then.
I also did it in parallel with the Language Transfer podcast, which I've mentioned on HN before but is an amazing resource for understanding how a new language works structurally and grammatically.
I did find Duolingo valuable, but purely for vocab.
Anki is great for learning (and allows the same spaced repetition), and AnkiWeb allows proper synchronization, but you gotta maintain and build your own deck (which itself aids you, but is a lot of work with things like pronunciation) or use another amateur's build deck. The power of Duolingo is that it reuses SVG artwork, and that the TTS engine plus voice sets work reasonably well.
1. It was funded by Duolingo but carried out by an outside team of academics. I can't judge how much this impacted their analysis.
2. No control group, which isn't very promising for the rest of the methodology.
3. More time spent using Duolingo did seem to result in higher improvement
4. Lot of dropouts, some participants excluded for taking up courses. They tried to control for 'outside resources' like watching movies in Spanish but it's unclear if that problem was really solved.
5. Novices of the language learned the most. Makes sense, as Duolingo focuses a lot of vocabulary and basic sentence construction
Personally I'm using Duoling to learn Italian, and I find that I am definitely learning, especially in terms of conversational phrases and vocabulary.
However, I've learned a few other languages by immersion, and this is going much, much slower. I'm 99% sure I would learn more by moving in with an Italian family or by consuming all my media in Italian, instead of Duolingo. For now, Duolingo is a nice compromise, and I accept that I'll learn a limited amount from it.
I think the correct conclusion from that study is not that Duolingo is good, but rather that it is not quite as bad as typical college language courses which are truly terrible.
If you're using duolingo, its probably because you dont have an opportunity anywhere else to use that language. In that circumstance, I don't know that duolingo is any worse than any other method of learning. However you're learning a language, however, you're going to have to get out and use it with other humans at some point.
Duolingo is a feel good tool - it makes you feel like your learning, perhaps quickly, without really learning anything.
For example, it uses multiple choice questions. Presented with a word (ananas), it then shows a few pictures (apples, bananas, pineapples and pears, for example). You incorrectly choose 'bananas', because really who wouldn't, and the pineapple lights up in green.
This method is really at odds with the 'best' ways to learn. If your recall is dependent on contextual cues, you will struggle away from that context.
Additionally, the volume really isn't that great. Like a lot of 'feel-good learning' platforms, duolingo encourages consistancy over volume. It's better to study for 10 minutes a day, than for 90 minutes every sunday. That's true, but if you want to learn a language (no small feat), then even 90 minutes a week is woefully inadequate. Being conversant in multiple languages is something to be proud of, and things to be proud of tend to take substantial effort.
Vocabulary acquisition is almost the perfect problem for spaced repetition, so use anki. There's a few strategies here: english word -> foreign word, picture -> foreign word, foreign word -> foreign synonym, foreign word -> english options, english sentence -> foreign sentence.
Pronunciation is difficult at first but gets easier. Accents are tricky - they're verbal gymnastics. Its easiest if you can find a native speaker of your target language, and you should just converse with them. Don't focus on minor details, because you usually pick these up naturally over time.
Listening is easy - listen to their radio, watch their TV. Similar with reading.
But there's one thing I haven't mentioned. Grammar. You do not need to learn foreign language grammar. You'll pick it up. How often have you studied English grammar? Unless you went out of your way, probably never. You learned some basic rules in your first few years of school, but everything else you picked up.
I used it to quickly get from close to zero to an A1 level - which enabled me to directly take an A2 levels course - instead of getting bored of a too easy one too quickly.
At least it saved me a hundred euros, but it's not a substitute for a real course and real interactions.
But later it talks helped in using more esoteric grammar that I never hear in small talk.
I'd suggest to use the website from time to time as it has some well done explanation to what is grammatically going on (the app is theory free)
I've been learning Portuguese to near-fluency over the past few years and I've started learning Russian intensively for about a year using a variety of resources but mainly language learning apps like Duolingo, Memrise, LingQ and a few others. I currently have a 800+ day streak on Duolingo: https://duome.eu/liofla
I find that Duolingo is the least useful of these apps. Its main advantage is that it's easy and a small commitment if you decide to do one or two lessons every day. For the rest I don't understand why it's so popular. Duolingo's main strength is that it makes you actually construct the language instead of just memorizing individual words or expressions but I'm not super impressed with that either, mainly because in my experience it's very, very common to have a sentence rejected because it didn't match the internal "regexp" used to validate it even though it's perfectly correct (and sometimes these mistakes linger for literally years despite being reported). I actually got a very mediocre result in the placement test for the "French for English speaker" tree despite being a native French speaker who's reasonably fluent in English, mainly because some perfectly correct answers were rejected by the system. That really destroys your confidence when that happens (both in yourself and in the app).
You have a forum to discuss these issue which contains some very valuable information but it's the worst forum software I've ever used bar none, it feels like a teenager's first PHP project in the early 2000's.
Oh and if you use the mobile app it'll have you build the language by selecting one full word at a time, which means that you usually won't have to thing about the conjugations/declensions and just vaguely remember what word means what. It's fine for English, not so much for Russian and its complex declensions and aspect system.
For vocabulary Anki and Memrise easily win because they only do that and they do it well. You can find podcasts aimed at learners of many languages to improve oral comprehension and LingQ is great to improve reading comprehension, although it's expensive for what it is and I feel like you could make a better clone of it in one weekend.
In general I haven't been really impressed by any of these language learning apps, it seems that they really lack the resources to do anything but the bare minimum. It's probably too niche to generate some real R&D.
So overall if you enjoy Duolingo then stick with it, but keep in mind that you probably won't get anywhere just using this app. If I had to recommend only one language learning app it would be Memrise because it's got some decent decks for many languages (including user-contributed ones) although of course you can't learn a language solely by memorizing the dictionary.
I'm a linguaphile with a lot of experience with Duolingo. I've joined
in the very first months of Duolingo. There are many problems with it:
* First and foremost, Duolingo isn't a product, it's a series of
products, which all have different characteristics. A person using
Duolingo on Android will have an experience completely different from
the experience of someone who uses iOS. And the difference is even more
drastic when comparing mobile with the desktop version. The desktop
version is sometimes three or four time more challenging.
* Even within one platform, the A/B testing has become so large-scale
that even two Android users might see a completely different product.
* Courses for “popular” languages get way more attention than the
“unpopular” ones. And the quality varies greatly.
* The ads they show are sometimes loud and obnoxious, NSFW, or straight
up scams. That doesn't happen that often, thankfully, but I've had my
portion of loud-as-hell game ads and borderline pornographic hentai game
ads. And before you say what people always say when it comes to NSFW
ads, Duolingo says that ads aren't personalised, so no, my search
history has nothing to do with them.
* Finally, it is indeed a good way to learn some basics and acquire
some basic vocabulary, but there is no way you'll get fluent with it.
A friend of mine has summed Duolingo up very well: it gives you
a feeling that you learn something, even if you don't progress at all.
That's just the stuff I could remember off the top of my head. It's
still nice to have it, but we still should remember that the service has
lots of issues, some of which could be eliminated, if the management
wanted to do so.
I'll second this. Another problem is the subscription, which IMHO, is overpriced for what you're getting. They also seem to spend an unusual amount of time redesigning already functional UI's.
I've been using it for more than a year to study Hungarian, pretty regularly : 2 lessons a day, pretty much no streak break.
It's pretty good for the first steps, and to get some vocabulary.
I can't follow a non-basic conversation, but I can pick up words from native speakers.
I'm now at the point where I should work on my own with more seriousness.
Overall, it's good to start learning a language, and to keep a regular practice, but it's not enough on its own to actually speak the language.
I've tried it multiple times (android) but I found it incredibly boring. So much time repeating the same basic words over and over for weeks. Real courses with live people are much more interesting, although expensive and challenging.
I've used Duolingo to learn Portuguese which is really close to Spanish (my native) for over a year, and it worked pretty well. Now I've done Dutch for over a year and it's a bit harder, I've taken 2 dutch courses as well, but I think the language is just hard for me. Though now I'm a somehow able to hold a small conversation.
I do 10 minutes per day, so it's fine by me, I usually waste more time in social media.
English is not my first language. We had 11 years of 4-7 hours of English a week, yet this was still not enough to get everyone to be decent at the language by the end of high school. The people who used English to play video games or chat online tended to be much better at it than others.
What I've realized is that experience teaches you to speak, read, and write a new language. You need a lot of experience to become good at it, because the language has to feel intuitive if you want to hold a conversation in it. There's the recommendation that if you want to learn a language you should immerse yourself in it (in real life). I think the reason why that works is because it forces you to figure out how to use the language and gives you an immense amount of experience in it.
I think that Duolingo is just another way to get slightly more experience in the language, but it's probably not going to be enough on its own. You're just not going to be spending hours every day on it to compete with language classes. From my experience, it was a nice way to quickly learn some of the basics of Japanese, but Duolingo really can't make you understand Kanji more easily than other methods.
I've used it near daily for 2 years to learn Spanish. Can confirm it actually helps, but only if you put in additional effort beyond the bare minimum the app requires. The app suggests that you speak out loud every sentence it presents you, and I usually do that, sometimes multiple times. I listen to them spoken by the app with my eyes closed to train my ear. I try to visualize the meanings behind words and sentences rather than just translating mentally to English. And I then seek out opportunities to practice Spanish in real life. I have the opportunity to travel to Latin America a few times a year, and I always go out of my way to use Spanish as much as possible there.
I had no Spanish background before Duolingo, or any spoken language other than English. I took Latin in HS, which helps somewhat with vocab and reading but not at all with speaking and listening. With solely Duolingo plus the additional practices I mention, I've gotten to the level where I can survive in a Spanish speaking country. I can't speak or listen well, my grammar isn't good, but I know enough to ask for what I want and understand basic information being told to me. I can also read enough to understand at least the key information from most written sources. I think I'll definitely need a more dedicated class and/or a longer full immersion experience to get towards my goal of being conversationally fluent, but Duolingo has been a great starting point.
Duolingo works reasonably well when used as a supplementary aid to learning a language. I've found that it has reasonable utility when combined with taking a physical class as a low-effort boredom filler for building familiarity with the vocabulary of a language.
What I find pressingly missing is any meaningful way of learning about the grammar of a language. When learning German there was absolutely nothing in the lessons which even hinted at the rules of tenses, cases, conjugations etc and the lack of that content stunts the possibility of building much more than a rote-learning knowledge of a language.
The app does seem to have a great gamification mechanic and is good at driving user engagement, so I guess they have that though.
The web version of Duolingo used to have detailed explanations for each lesson that explained a lot of grammar, they seems to have disappeared recently
I speak English (native), French and German. I’m currently learning Dutch with Duolingo. I’ve only used the iPhone app for about 2 months at the free tier.
I’ve already learned similar languages the traditional way and I go to the netherlands at least 1-2x year for work.
Personally, I feel I get enough value from Duolingo to make it worth my time. But I admit I’m a particular use case.
IMHO it is not was way to learn language if it is one’s only tool. But then again neither is only classroom instruction or only books or only tv. I think that is where people get hung up on critiquing it.
And yes, I've dabbled enough to know that I could learn Brazillian Portuguese and try to convert, but I've also dabbled enough to know that to do so would be really hard, and leave behind some undesirable remnants.
> I've also dabbled enough to know that to do so would be really hard.
No, that's not true. Learn Brazilian Portuguese and you'll be 95% there. The major differences are in the voçe vs. tu (second-person singular pronoun). There are other differences (mainly pronunciation), but the core of the language is the same.
Source: I'm from Argentina and learned Brazilian (travel often). I have a few Portuguese friends and I have 0 issues talking and understanding them.
Talking with and understanding them is not the remnant to which I refer. I'm not sure I can explain it, but I've found in generally, converting a "near enough" version to a "proper" version is something I find difficult. Experience of my own learning tells me that if I start with Brazillian PT then I will forever be labelled as a speaker of Brazillian PT, even if I try hard to leverage it across into European PT.
Understanding and being understood is not the entire point here. Having said that, I'm finding virtually no resources for learning Euro-PT, so I'm likely to have to resign myself to always being thought of as a speaker of Br-PT, despite living a few hundred kms from Portugal.
(BTW - thanks for the response, it is a useful data point)
Yes. But I live in Portugal now and I don't want to speak like a Brazilian. I want to speak like a European Portuguese person. Duolingo should clearly label their "Portuguese" as "Brazilian Portuguese" as the pronunciation differences between the languages are massive and there are grammar and vocabulary differences.
I’ve been using Duolingo to restore my limited knowledge of Japanese and worked pretty well. I managed to keep a two month every-day habit, but collapsed as soon as I skipped few days — there’s something to learn about too much aggressive gamification here.
Nonetheless I suggested my mother and grandmother to start using it to get better at their english and they’ve been using it every single day since 8 months ago — and I’ve seen incredible results. They start to understand basic online content and they are asking me to translate content less and less every week.
I'm hesitant to use duolingo, after seeing the amount of funding rounds and VC capital they have taken. I would steer clear of using it.
Are there any free and open source alternatives to duolingo out there? I wouldn't want to use something that could shutdown after an 'incredible journey'.
I don't think there is anything comparable OSS wise. Creating such an alternative wouldn't only be very difficutl tech-wise, but someone would also have to create the content.
Anyway, I've been using Duolingo since ca. 2014 and it didn't shutdown... Plus, it's free - what do you have to lose? Your progress on some language learning app? You won't lose the knowledge gained.
Plus being a betrayal of their user-base which is heavily weighted towards young and impressionable learners. And, by the way, a betrayal of their initial promise, that they would never do advertising (expressly stated in the first promo video they released, which is still on YouTube).
You ever hear of a thing called opportunity cost? Some people interested in learning a foreign language actually have a compelling desire to learn said language.
What you have to lose is your time spend. At least you'll lose time due to the gamification and the ads.
Anki is useful for spaced repetition, but setting up a deck is going to take effort and requires knowledge (building a deck is useful by itself though).
I was using duolingo recently for Spanish. Unfortunately, I got dropped into the "hearts" A/B test and it destroyed my experience.
Basically you get four hearts every few hours. Every mistake takes away a heart. If you lose all your hearts you can no longer play. But, you can do _review_ lessons (that have nothing to do with the content your currently learning) and regain 2 more hearts.
It basically became a grind situation and I stopped using it. I've heard the webapp doesn't use hearts, but I refuse to support them after that.
And, yes, I was planning to buy a program to learn Spanish and had been using duolingo for 2 weeks prior to getting the hearts.
I did some reading on reddit and it appears this is related to some VC funding and attempt to make money. I understand they need to make money, but this felt like a very bait and switch slap in the face kind of thing.
The whole experience kind of soured me on learning for now and I haven't been spending as much time on it since.
I did find clozemaster which seems to be a similar, and honestly better, app. I also have a textbook and workbook pair. And, fortunately, have a lot of native Spanish speakers around me.
Very wise. You can see the level of VC control in the company and how they are only continuing to increase their influence as they head towards the IPO. The VC funders want their monetization payout. It's somewhat alarming considering the userbase is mostly teenagers - and younger.
It's true that it's tough to learn a language from scratch with just Duolingo - only language classes truly worked for me. But as a means of review or for giving my friends who would probably never learn the language properly a leg up, it's invaluable. Rosetta Stone used to be a $700 investment and this made it obsolete.
I learned Mandarin and German pre-Duolingo, both to reasonable levels of proficiency.
I've messed around with Duolingo a few times, just to get a feel for how useful it is. I came away feeling that it was about the same level as flashcard vocabulary learning - useful, but only a small part of the pie.
Vocabulary is a necessary, but not sufficient, condition to be able to carry on a conversation. Mandarin in particular is more difficult, since you need to grind through hours and hour writing/rote learning characters to improve your vocabulary.
The one thing that noticeably accelerated my proficiency was having an actual conversation with a native speaker (host family/tutor/etc). That's not something that Duolingo really replicates.
Self-plug - coincidentally, earlier today I just released a small tool for Mandarin learners to practise their tone recognition! I've been working on a larger app and this seemed like something that would have helped me when I first started out. App is currently live @ https://fluent.academy - I'll be introducing more functionality for Mandarin learners in the near future, so would really love to hear from students of any level.
Thanks - the voice is actually synthesized, not recorded, and slowing the pace occasionally produces some artifacts. I'll double check the ones you mentioned!
Out of curiosity, what mobile issues are you encountering? This is actually a Flutter web app, which isn't officially production-ready yet, so I'm keen to provide feedback to the Flutter team if something's going wonky.
I agree. Duolingo is great for learning vocabulary, but if you really want to learn a language you need to supplement this with listening to the language as well as speaking it.
I've been learning French and have been using Duolingo for vocab, but then I listen to a French-teaching podcast as well as trying to work on reading/speaking it.
I'm not at the stage where I can listen to it since it's such a fast language. I've been reading LeMonde newspaper and trying to decipher it that way. I live in Canada but the French here is quite different in each region; Quebec, Acadian, Métis. Duolingo French course is France French not Canadian French I think it's standard French aka Parisan.
There are podcasts for learning such as "Coffee Break French". Basically they start with very basic french and build up. This can be good for starting to learn to listen to french.
Regarding the differences between Canadian and France French, are there that many differences that it would make it difficult to understand someone from another region?
Native French here. One day I was in a place with 4 people from Québec speaking with each other. I could understand perfectly three of them, then it took me 15 minutes to understand the last one was speaking French too and not a foreign language. So, accent could vary quite a bit between region and people but most of the time it’s similar enough that there is no problem.
In addition to what others have said I can only offer anecdotal evidence.
I have friends and former coworkers some from Quebec, some from New Brunswick all fluently bilingual. We've all worked with people from France on projects and my Acadian friends remarked they couldn't understand a word the people from France said.
Although the reverse of that a friend went to northern France on vacation. When he was there he said everyone asked him where he was from he told them New Brunswick Canada. From what I understand Acadian (French dialect of New Brunswick, Canada) is older and closer to northern France than the Quebec dialect. I'm not sure if my friend meant the regional dialect or standard Parisian French, I'm pretty sure he meant the former.
And this may be totally wrong but I believe the southern France dialect (Occitan) is quite different than the north. I've had to call Monaco (I know it's not France) and I was surprised at just the accent of the French person speaking English. I've been around French all my life but southern France to me sounds like stereotypical cartoon French like Pepe LePew.
The entire history of standard French for their language is quite interesting. It's not been accepted for as long as you may think and there are many dialects.
edit:
I think Canadian French has a lot of slang not used in France which makes it hard to understand. Kind of like a Cockney from England coming to the US, sure you can understand his language but not the meaning the way he uses words.
What if you don't need listening or speaking skills, only reading? In your opinion is Duolingo a good fit for someone who only needs to learn to read the language?
Not the gp, but I used Duolingo quite a bit, and recently learned Spanish to an upper intermediate level. In my opinion, Duolingo is just as weak with pure reading as with spoken language. You'll learn some vocabulary and be introduced to ideas, and that's valuable, but not enough. Just like spoken, to learn to read, you have to read a bunch of stuff you can understand. I jumped straight from Duolingo to children and young adult books, and it was a difficult transition with lots of intensive looking-up-every-other-word study, but, I mean, you do that long enough, and eventually you're just reading.
I have been learning Mandarin casually for a few years. A while back I went all the way through Duolingo. I picked up some new vocabulary, but it was probably the least effective study method I used.
With that said, it was reasonably fun, and I did stick with it for a while due to the gamification, so maybe there's hope for the feature. Just getting people to study a foreign language regularly is probably a good step.
I also think that they have implemented some more advanced learning features with a few languages.
"...the same level as flashcard vocabulary learning"
How exactly were you using it? It speaks to you, which is invaluable for developing an ear for the language. It forces you to translate both ways, which helps with reading and writing immensely. And optionally you can even speak answers in the foreign language to practice pronunciation (such that your phone can correctly transcribe what you're saying).
It's not exactly an immersive language course, but it is vastly more useful than flash cards.
I know a lot of people like to speed through the tree and only complete the first level, which I suppose is a more vocab-focused way of using the app. But if you go deeper on each level it forces you to grapple with grammar a lot more.
> I know a lot of people like to speed through the tree and only complete the first level, which I suppose is a more vocab-focused way of using the app. But if you go deeper on each level it forces you to grapple with grammar a lot more.
Probably a fair point - as mentioned I wasn't really using it to learn the language, rather to get a feel for "how useful would this have been when I was learning". So I admit I didn't dive into the later levels.
Asked me to sign in with Google. Couldn't click behind. Closed window.
Come on, don't make me sign in everywhere. This is just email harvesting. I get if you want to sell a service, but then I need to know what you are providing. If it is good I'll pay for it (as someone trying to learn Mandarin. There just aren't good tools out there). But please don't just blindly collect user data.
I don't know if this is intentional or not, but if you click the back button when the google login modal appears it goes away and you can continue to the rest of the buttons. I ran through a few of the questions and didn't get another sign up prompt.
Totally understand - at the moment I was just asking for registration so I can stay in touch with people who might be interested as more features get added or when the larger app is released.
In the meantime, I've enabled e-mail registration without verification, so feel free to signup with a throwaway :)
Okay, I did. Now I have to say that I do not find this helpful. The audio is very synthetic and has audio distortions. 三 had a weird glitch in it. Have you tried using one of Bidu's models?
Also, in counter part to Duolinguo, I suggest at checking out Hello Chinese, and using that as the metric to beat (it is popular among Mandarin learners).
Thanks! Recognizing and remembering tones is one of the things I struggle with most, so I'd use this a lot.
A couple of comments: for 水果, it only accepts shŭi gŭo, not shúi gŭo. I know that is technically correct, but the voice synthesis sounds it out with the sandhi. More a question of philosophy I guess, but thought it was worth mentioning.
Also, it might be nice to have an option that hides the characters, so you're forced to rely on hearing rather than just knowing that e.g. 多 is dūo.
Looking forward to seeing what else you do with it!
Personally, I've found Duolingo to be almost universally worse than any other language product. It seems to have become successful and widely-known purely from marketing. It is the current generation's Rosetta Stone.
In general, I recommend Glossika or Pimsleur over Duolingo. Depending on the language you're learning, there are also more niche sites that do a far better job; Turkish Tea Time or RealPolish.pl are some good examples.
Interesting. I strongly dislike Duolingo (because it seems very focused on passive recall, aka multiple choice), but think pretty highly of Rosetta Stone.
I haven't looked at either of Glossika or Pimsleur, but Glossika's page title is "AI-based effective language learning" which immediately reeks of marketing blather. Why do you recommend them?
Yeah, their new branding is a little strange. But the original Glossika method (which the new software still essentially does) is to teach a language only via phrases - not grammar trees, isolated vocabulary words, or anything else.
I’ve found this approach to be more fruitful than the abstract one most books and programs take.
> the original Glossika method (which the new software still essentially does) is to teach a language only via phrases
Based on the screenshots, it looks like they maintain a "memory level" for each phrase and presumably do spaced repetition to keep that level above a certain target. Am I right in assuming that if you fail a review because you forgot what one specific word means, you'll get the exact same sentence next time, rather than a different sentence using the same word?
I have a system for personal use that does the latter, but I haven't yet found a good interface for indicating why I got the sentence wrong. My current implementation is to have a huge list of checkboxes for every aspect of each word (dictionary form, spelling, pronunciation, etc.) that could potentially be forgotten independently.
> Am I right in assuming that if you fail a review because you forgot what one specific word means, you'll get the exact same sentence next time, rather than a different sentence using the same word?
Correct. It’s just based on the sentence, not the words within the sentence.
How's it built? What languages do you use this with? My biggest concern before starting something like this is the I've thought about scripting the generation of example sentences for myself, but I'm unsure I'd be able to do it in a grammatical way (since I don't know the language, a catch-22).
When I started building it for Japanese, I knew a little (half-forgotten hiragana and katakana, also many kanji because I speak Mandarin), but most of the language-specific features come from existing tools I've just glued together.
I have two main sources of example sentences: 52114 from Tatoeba https://tatoeba.org , most of which have translations, and 114591 extracted from public-domain ebooks from Aozora Bunko https://www.aozora.gr.jp/ , some of which use rather archaic language.
I process each sentence using Kuromoji https://github.com/atilika/kuromoji to segment it into individual words and get details like part-of-speech tags, dictionary forms (lemmas) and pronunciation for each one. Then I add them to a SQLite database.
Some sentences on Tatoeba have audio recordings contributed by native speakers, but there aren't so many for Japanese, so I also use Open JTalk http://open-jtalk.sourceforge.net/ to get a robotic pronunciation. That pronunciation may differ from the one determined by Kuromoji, which I use as a consistency check.
I have tried various criteria for determining what to learn next. One that worked well is to keep track of the least frequent unknown detail of each sentence (e.g. the spelling of a particular word) and pick the sentence where that frequency is highest. That way, sentences with common words come first. Lately I've modified it a bit to also consider how many other sentences have the same least frequent unknown detail, so I always have multiple different examples available to help me generalize.
Reviews are scheduled for each detail based on the time since the last review and the time when I last had to relearn it. I've been collecting data to later tune the review schedule, but right now it just doubles the interval between reviews every time I remember correctly.
I used to just pick a random sentence for each detail with a scheduled review, but that would sometimes randomly schedule the same sentence often enough to be noticeable, so I've changed it to prefer sentences I've never seen or not seen in a long time.
The two kinds of review are dictation, where I listen to the recording and write it down on paper, and reading, where I read the sentence aloud. In each case, I manually check whether I made any mistakes and indicate them by unticking a checkbox (as mentioned in my previous comment). Those details will then be scheduled for review more frequently.
After a year of daily use, there are now 35611 sentences in my database which could potentially be scheduled for review (which means that, theoretically, I should be able to understand them), 21.3% of the total. The ebook which I should understand the best (35% of all sentences) is the Japanese translation of Karel Čapek's play "Rossum's Universal Robots" https://www.aozora.gr.jp/cards/001236/files/46345_23174.html 35% understanding is ... not enough to allow me to read it from beginning to end, but occasionally I get a nice sentence e.g. about the worldwide robot uprising.
I think it should be possible to adapt the system to work for other languages, but it would require significant refactoring to remove the assumption that everything is in Japanese.
Thanks for the detailed reply! I'd be interested in reading more technical details about the whole pipeline if you have written about it.
One question that sticks out in particular: how do you actually perform the reviews? Did you create a custom GUI for it or are you using an off-the-shelf software?
My comment above was the most I've ever written about how it works, except for one very long comment in my database migration code, which I had to write to remember how it's supposed to work.
I created the GUI in Qt. It's not beautiful, but gets the job done. I decided to finally open-source my WIP and put a screenshot in the README: https://github.com/Yorwba/alphabet-soup
If you don't care about all the Japanese-specific stuff, you could try running make data/tatoeba_sentences_XXX.csv where XXX is the ISO 639 three-letter code of the language you're interested in. That collects all the sentences in that language contributed to Tatoeba by native speakers and puts them into a TSV file.
Thank for your recommendation of Glossika I've never heard of it before but it looks good. I'm a beginner I've been using Pimsluer (for Mandarin) for a while while.
Although it's not perfect. I prefer Pimsluer over Micheal Thomas Method, which I've done previously. I'm just using it as a source of comprehensive input to develope my ability to able to understand spoken Mandarin (which MTM was awful for).
I'm sure I could use other programs similarly. Though Pimsluer probably has the most material. That might not matter if I get good enough to use films without subtitles and podcasts for my sources.
Glossika looks like it's geared more to how I'm trying to learn. I'll probably give it a try at some point.
I enjoy Pimsleur, especially for learning bits of a language before travelling somewhere. It's fun to listen and practice with my partner. We both sound silly, and we can encourage each other to improve our pronunciation.
I think that's the hard part about learning languages: it works way better if you don't do it alone, and that's why apps can be pretty limiting.
I think Duolingo is fine if you're looking to learn a language that's really close to yours. I know French, took up Spanish and within a month of practice I could cobble simple sentences in a casual conversation or text my Spanish-speaking friends. The approach where you just "figure out" from trial and error works well when your source and target languages kind of work the same way so you can just feel your way through it. On the other hand, I suspect that if I wanted to learn, say, Russian or MSA, I would probably be better off with actual formal courses.
I have only given Rosetta Stone and Pimsleur a cursory attempt but they seem awfully slow and expensive for what they do. Duolingo, for all its flaws, is free and fun.
As for your "purely from marketing" comment - Duolingo solely relied on word of mouth for all of its early phase, apart from a single TEDtalk. It's true it may not be as good as many alternatives, but as far as the free and fun factor goes - it is simply unmatched.
To be fair. Rosetta Stone got it's "reputation" because of it's insane price point comparative to what you learnt. Duolingo on the other hand is completely free.
There are only marginal benefits to Duolingo plus that can't be obtained with a proper adblocker. Many people who choose to subscribe admit they only do so to support the company.
I see duolingo as a kind of modern equivalent to the language textbooks we had in school. You aren't going to seriously learn a language using it, but it gives you enough of a taste that you might be motivated to decide to pick up a language after using it.
I do think that using duolingo too long is an inefficient way to learn a language even at beginner level, I wouldn't use it for more than a few weeks before graduating to a SRS like Anki.
For intermediate to advanced I really love clozemaster. It's especially good for learning grammar, you probably want to supplement it with some pure vocabulary in an SRS.
I used DuoLingo for quite some time before using Anki. My experience with Anki was much better.
My experience was that I was much better at recalling vocabulary from using Anki than DL.
Possibly because the "yes/no" to DL is at a phrase/sentence level rather than per-word.
Without putting effort in, DL makes it easy to recall what the DuoLingo-defined accepted translation of phrases is. It's easy to fall into the habit of 'gaming' it and just remembering the answer in a very shallow sense.
With Anki, using flashcards of words, the 'effort' you need to put in is being honest about whether you recalled the word.
I felt that when I started using Anki, I was much better at answering DL.
I have a 2201 day streak playing Duolingo. I love Duolingo. I have taken Spanish and French that whole time and German for a few years.
I find Duolingo to be an extremely fun game. My wife and I really enjoy trying to speak other languages and even more so now that we have talking children. I can speak passable Spanish and can string together broken sentences in French to be effective (they get what I mean) and hilarious to native speakers.
Duolingo has never purely taught me a language. If my wife didn't know Spanish and learning French like me, I would have never been able to practice enough in the real world to know.
I enjoy using Duolingo immensely. For me it is a fun way to toy around with a hobby I've had for years without committing myself to dedicated training costs and time.
That last sentence is exactly what I like about Duolingo. 716 day streak here. Without Duolingo, I would not have commited to at least some minutes spending with a foreign language. Now I can at least give directions in French. I managed to have 8 years of French at school without being able to do that in the end.
I’ve been working on my own version of “DuoLingo” for iOS for 10 years now; longer than DuoLingo. I’ve never gained much traction. I’m looking for other ideas to help language learners.
Recently, I’ve tried writing smaller apps to supplement what Duo provides because they have a great product and I’ll never compete directly as a solo developer.
Anyway, any other small app ideas that people would like to see implemented? Game ideas that help with the necessary repetition, for example?
One idea would be to use a wide variety of pictures for learning a particular word. If I am learning what the word for room is, if the picture is always the same picture of a bedroom, my mind will be building an association with bedroom instead of room in general.
Also as someone trying to pick back up on my Japanese after letting it slide for a decade (well to be honest I never had more than enough to do greetings and basic restaurant ordering), many kana/kanji matching apps leave a lot to be desired. Katakana is a personal weak point and the apps I've tried all limit the extent of matches to picking one of the matching 3 or 4. Process of elimination makes it so that I can get 100% of these quizzes despite only knowing 80% of the material. Something that would allow customizing 10 or 20 possible answers to pick from would be nice. But I don't know if this would have any bearing on other languages to be worth developing (and there might be an existing app I just haven't found yet).
> wide variety of pictures for learning a particular word
That idea doesn't scale. Much simpler to look at definitions and example sentences, especially for non-obvious words like abstract ideas, context-dependent words, and particles.
> Katakana is a personal weak point
Just use flash cards. It's rote memorization, so there's no substitute for time and repetition. If you're able to commit an hour each day, you should be done by the end of the week.
The problem with looking at words is that you are learning language to language translation instead of language to thought. When I think of "pencil", I have a schema of a pencil appear in my mind. When I see enpitsu (well, the hiragana or kanji for it), if I have to think "pencil" to bring up the schema of a pencil that is an extra layer of thought that slows one down. With more abstract concepts there might now be a better solution, though with more abstract concepts you also have greater issues with things not being one to one translations.
>Just use flash cards.
That is effectively what the apps are mimicking and what I am using. What I'm bringing up in a desired feature I haven't found (granted, I've only checked out a dozen or so apps which is worlds apart from serious market research). As for why I use an app over flash cards? Phone is more convenient and always on me. Also, other than a stack of the most common kanji, flash cards don't really scale in a portable format.
But I probably should buy a deck including kanji for at home usage.
Sure, the ideal end state is that you know the Japanese word intuitively without English or another language as an intermediate step. But starting from a totally unfamiliar word, dictionary definitions and example sentences are both fast to access and easy to copy down. (For a sufficiently advanced learner, I would encourage using Japanese language dictionaries instead of Japanese-English ones.)
And I'm only recommending flash cards and other rote tools for kana, because recalling them instantly is both important and relatively low-effort. I don't have a strong opinion on how to learn vocabulary, other than to prioritize based on actual usage: media you read/listen to, conversations, etc.
I'm not sure exactly what you're after, but the Hiragana Pro/Katakana Pro apps (android only, I believe) have two practice modes: choice from three pronunciations as you mentioned, as well as typing romaji with no hints. The latter makes it abundantly clear which kana I'm shaky on.
They're also each a dollar to unlock voiced/combined readings and remove ads. My pihole did block the ads while I was just trying them out.
>I enjoy using Duolingo immensely. For me it is a fun way to toy around with a hobby I've had for years without committing myself to dedicated training costs and time.
My exact experience. Been wanting to learn Dutch for years but I just can't invest the time and effort. I've got a 300 day streak on Duolingo now and I'm close to A2 level and I can already participate in some online conversations and understand a great deal. It is language learning for "free", I'm not talking about monetary cost here. I wanted to learn a new language but didn't want to put the effort in. It feels like I'm learning for free, without the effort which is exactly what my lazy ass wanted.
If I were in a hurry, or wanted to put in serious effort, it probably wouldn't be the most efficient way. But I'm not in a rush. English is my second language and my formal training brought me to B1 level at most. The rest I learned through exposure online over the years (and got an almost perfect TOEFL score a decade ago) - I assume the same will happen with Dutch too - I got a foot in the door thanks to Duolingo. I don't care if it takes a couple more years compared to another method that I'd probably never consistently use.
No I haven't. That feature came out years after I started.
I do have tons of twilio/glitch based reminders that I moved from crons on a VPS in 2017. I am aided by technology to remember, but I usually do it early in the morning.
I had a near 1000 day streak when I had a medical procedure and was using some heavy medication. In this time, I lost my streak by 1-2 minutes. Ever since, I quit. It felt like a choir, so I was happy it was over. That said, it allowed me to learn a lot of Spanish. I also found out Spanish is a relatively easy language (while I always detested French as I found it very difficult). A while after I quit (my partner still plays it) they added annoying ads and subscription.
I had something similar happen — when my streak ended Duolingo asked me for $38 to restore it. It made me feel manipulated and sick to my stomach. I didn’t use it again for three months. Now I’ve turned off the leagues and try not to get sucked into the gamification. I’m a little sad because I used to love it. I wish there was a way to turn off the streak and XP and everything else.
2201 is really impressive streak. My current is just 355!
What I like about Duolingo is how casually it approaches you to make consistent efforts every day, without overexerting you to train, like many fitness apps usually do.
> I have a 2201 day streak playing Duolingo. I love Duolingo. I have taken Spanish and French that whole time and German for a few years.
> I can speak passable Spanish and can string together broken sentences in French to be effective
Things like this worry me. If we look at how long Spanish and French take, it is about 600hrs[0] (probably less because specifically French is really close to English and Spanish is close to French). If you only had 20 minutes a day you should be reasonably fluent. But if your wife and you are talking, I'm guessing you average more than that.
This is a game that I enjoy playing. I have never tracked how long I take. Many days, I do only a single lesson. That's less than 5 minutes, sometimes less than 3.
I'm not trying to be fluent. I'm not really _trying_ to achieve anything. The closest I have to a goal is that I enjoy having another language's vocabulary that I can use when speaking around my children.
Such is the nature of language learning. It's not really sexy or fun to study vocab or shadow along to something you're watching for the 10th time. I see language learning exactly the same as working out- it's just a lot of consistent hard work over years. ;-)
I think there’s a big difference with “practicing” sentences with your wife where neither has a good accent or the best ability to spot mistakes or knowledge of the word the other person doesn’t know (though I have no idea how much or how good the GP’s practice was), and practice with a native speaker (or even better an actual teacher too) who can quickly tell you about mistakes and unknown words to give faster feedback and more deliberate focused practice. I’m not at all surprised that GP’s Spanish is much better than their French.
The issue here is that the GGP has been doing this for 6 years. If this was a year of practice I wouldn't be surprised. But at 6 years? I think they hit a plateau.
Duolingo is a learning game. It should not be treated as more than that. The dream is that all we need to do is play to learn a skill but Duoling has not found the secret sauce yet so it will not teach you a new language all by itself. You need to use it along with a teaching system.
What's a bit distubring is that it is being taunted as a complete system for beginners.
I think it's a disservice to new solo learners since eventually people give up and have not learned as much as they could have given the time and effort put forward.
I think Duolingo has a place but it should be an aside to a proven learning system not the only tool.
I would love one of these vocabulary building apps if it came in the form of an Android Auto application that I could use while driving my 45 minute commute. Doesn't seem to exist, though.
I learned couple of languages with the help of Duolingo. They claim t be the fastest way to learn a language.
I recently realized is only valid if you don't want to put the work in. People who are not willing to put serious effort into learning a new language will find Duolingo useful.
It's also useful if you come from similar language. E.g. you speak Italian and you want to learn Spanish (latin langauge). Or you're Czech and you want to learn Polish (slavic language, same grammar and compsition)
But if you are serious about learning a language, Duolingo will only slow you down. You learn conversational stuff and progress often with big holes in the knowledge and understanding the core principles and structures of the language.
For serious learners of a new language, keen to progress properly and fast Duolingo should be avoided.
Duolingo works by means of repetition. The user can learn as much as they want. You need to define "serious learning" and "serious effort" if you want to make a well written critique. Which alternatives do you suggest? How much do they cost? How much time per month would you spend on these? Versus Duolingo? What is/are your end goal(s). In how many months would you reach your end goal(s)?
Serious effort is when you drill the grammar and memorise tense and forms of words. You'll have a base to build on.
Take English for example, difference between past simple vs. past continuous is easier explained once than randomly coming across on Duolingo never fully comprehending the principle. Therefore it's faster.
And form someone coming from non-Germanic language this will take time to comprehend, kind of like when you first learned some math concepts as a child, some took a while to click.
Now take words and their past forms, give, gave, given.
In order to use it in a natural way, you first have to drill and memorise many words ... not possible on Duolingo as one lesson you work with a bunch of "gave" and suddenly several lessons you use "given" ...
Yes, you can learn language in way that you go China and you're able to make primitive conversations. I know a lot of people speaking French that they learned conversational way and all of them struggle with grammar and despite being able to talk and communicate fluently their language is not enough for professional level or academic level.
E.g. my girlfriend and I speak English, a language native to neither of us ... we can live a complete life together and we can talk but to a native speaker her mixing of tenses and grammar would sound like nonsense.
This is the case with many of my foreign friends who learned English in a conversational way.
Duolingo does the best what it's designed for. It's a way to learn for the lazy, so it is a game.
This is a very arbitrary definition of serious effort. I learned one language to the point of being reasonably fluent without ever "drilling". I just listened and spoke in class
You also conveniently ignore the countless people (including adults) who reach fluency without even studying, instead just picking it up from immersion: work, media, etc.
Duolingo is a specific tool. Like any tool, it's best for specific uses. Your judging its users as "lazy" is both baseless and unhelpful
What is "fluent" to you? I highly doubt anyone has ever become fluent in any language from a class. Classes are always behind, and teach textbook variants of a language. Useful, but won't make you "fluent".
Immersion is still studying. Digesting actual native material ranges from hard to very difficult (i.e. dramas vs comedy shows, etc). Immersion + studying textbooks is 1000x better than gamified sources like duolingo, particularly for Chinese/Korean/Japanese.
Also without knowing your native language and target language, the comparison is not useful. Spanish is quite easy for English speakers; good luck learning Cantonese without studying [for a very long time].
You come off attached to your existing conclusions: All duolingo users are lazy, and nobody can ever become fluent in a language from a class. You're free to believe these things if you want, but it will prevent us from having a constructive discussion. Cheers
Nope. Not everyone has a lot of time to dedicate to studying a language, which is fine. But duolingo is never going to give you any meaningful fluency. It's like saying "I want to get ripped by only doing body weight exercises for 5 mins/day". If your goal is just to say "Hi my name is X, where is the bathroom", then by all means use duolingo.
> nobody can ever become fluent in a language from a class
No class will ever teach you everything about a language, partly because there is way too much to teach, and partly because languages change all the time and classes can't keep up. Instead, classes give you a starting point to start reaching fluency, which begins once you start consuming native material and having conversations with real native speakers. Classes don't teach you slang, pop culture, casual expressions, and so on.
Fluency is not making simple paragraphs or having a very basic, limited in scope & topic 10 min conversation. Language learning literature and other resources have lowered the bar for what "fluency" is, because nobody can sell a book or other resource by saying "use this book and learn french in 3-5 years". There is a lot of snake oil in this area, but people also want to believe in this notion that they can fluently speak a language with minimal effort.
Memrise is better as its more structured. But it's only as a supplement. Duolingo is also a good supplement, not the primary tool. Best to take some class and drill textbooks with fill-in exercises. Then books with short texts on different topics - highlight words, memorize. Then find ways to practice in real life.
Yes people say they used Duolingo to being close to fluent, most people who speak very bad grammar and mix tenses but people don't correct them mush as they get the point. So most are unaware. I speak it from first hand experience knowing about 10 people like that.
> Best to take some class and drill textbooks with fill-in exercises.
I disagree with your analysis. Most people aren't trying to be speaking super gramatically correct sentences. Most are learning to speak passable sentences.
Most important thing while learning a language is to have fun, so you can keep coming back to it. Duolingo makes it fun and interactive. Doing endless grammer textbook excerises /drills takes enormous will power that casual learners don't have.
That was exactly my point. And Duolingo is the perfect tool for such people ... for those who want to master the language and have the will Duolingo will slow you down.
Also a lot of people speak here from English native point of view ... yeah it doesn't matter you don't speak proper grammar when in Spain once in a while or while nomading. It does matter when you're foreigner and want to build a career or be successful in business/society in the said country.
Glossika (basically a modern pimsleur) is the best app that I've found for practicing a language.
If you actually want to learn a language?
- find one or three good textbooks and follow them completely with exercises
- find a language exchange, but it won't be helpful for speaking until you know 200-300 words and can speak a little. exchanges can be good for making friends, though. they're great for getting feedback on your pronunciation, speaking, etc, but always take things with a grain of salt. native speakers are not always good teachers.
- find graded readers or native content that is catered to early, medium, advanced levels, etc. immersion is really important
- use Anki for spaced repetition, to memorize vocabulary (that you found from native content like tv shows / books / etc)
- make friends thru language exchanges or apps like Hello Talk / Tandem
- in general try to get as much input and output as possible. people usually neglect speaking and listening.
- finding a weekly tutor can help you iron out personal issues, particularly with pronunciation
Doing an app for a few mins a day is never going to teach you a language to a "fluent" degree, despite what marketing they have. Learning a language requires serious study and practice- speaking (and writing, etc) are muscles that you have to train over time by interacting with real people.
I don't consider making a few contrived sentences like "my name is X, i do y for a living, where is the bathroom" to be fluent. Not at all. That's called being barely functional. Real fluency depends on your native language and target language, but requires years and years of work. It's not sexy like the idea of using an app and magically becoming "fluent" in a language.
Well, I put hours and hours into French in high school (at times, with serious effort) with very little result. My grades for French were awful.
Spanish with Duolingo, was at least pleasant. If I watch a series like La Casa de Papel or Orange Is The New Black I understand some words (more than something in French).
How would it have been vice versa for me? I will never know.
I like Duolingo because it's easy to get going and quick to take lessons during my free time.
It's easy enough to get going and hard enough to feel like you're doing something but not so much you give up.
The points system, scoreboard, and email prompts are a good way to encourage you without being annoying. The ranked scoreboard really has me wanting to stay there and compete to stay in the top 10 each week.
One thing I miss is the speaking option. Not all languages offer it but some did and that option seems to be missing or at least on my device it is.
Another problem is some languages are hit and miss with pronunciations if there are few native speakers. I know there are computer generated voices but sometimes there is no sample of the words at all (Irish being one example).
Anything that gets people interested in languages is a good thing. Duolingo may not be perfect but it's a popular and easy to use app. We unilingual English speakers need to get to with the times and learn at least one second language, maybe even more.
I've been using the voice-to-text included with the google keyboard on android to practice speech on duo. I'd be interested to hear from a native speaker how accurate the voice-to-text model is but I find it fun getting a computer to understand me in a different language. It can be pretty unforgiving for certain sounds that aren't in my native language so I feel like I really have to nail the pronunciation.
This is really Duolingo in a nutshell... the complete gamification of language learning. You don’t actually have to learn anything as long as you “feel” like you’re doing something.
> Anything that gets people interested in languages is a good thing.
I don’t share this view. If you want to really commit to learning a language for some reason, that is fine and great and there are now countless resources to do so, getting language partners is easier than ever... but it’s not a moral imperative, and if you’re just spending time on activities that have no hope of leading to fluency it really is just a mastubatory exercise (albeit one that is more appropriate to talk about in mixed company), or basically a game... which there is also nothing wrong with!! I masturbate.. but then again I don’t lord it over everyone as some moral imperative.
The more serious point: if Duolingo gets your rocks off, great, but as language learning tool it’s pretty poor. Sure you have to work to learn a language, but if you’re going to commit to that effort there are staggeringly better options, which given all the hype is a bit disappointing.
Duolingo in a nutshell... the complete gamification of language learning...You don’t actually have to learn anything as long as you “feel” like you’re doing something.
> Anything that gets people interested in languages is a good thing.
Whatever. If you want to really commit to learning a language for some reason, that is fine and great and there are now countless resources to do so, getting language partners is easier than ever... but it’s not a moral imperative, and if you’re just spending time on activities that have no hope of leading to fluency it really is just a mastubatory exercise (albeit one that is more appropriate to talk about in mixed company), or basically a game... which there is also nothing wrong with!! I masturbate.. but then again I don’t lord it over everyone as some moral imperative.
I've used Duolingo and like it for learning grammar and vocabulary. Why would someone pay for it though?
I would have been prepared to pay but I don't see why I would. The only paid features I see are to remove ads (which you rarely see compared to how long you use the app for), offline use (which isn't an issue if you have good mobile internet) and to skip to harder stages (which seems counterproductive).
In terms of making an income, it feels they give far too much away for free. Do they make most of their money from the ads? Or so many people use it for free that it's okay if only a low percent convert?
Ultimately, I primarily opted for the paid version of DuoLingo to support their mission of free language learning for all. The ad removal and offline use (great for planes and while traveling) were secondary reasons.
I opted for the paid version for exactly the same reason, but I was sorely disappointed to learn that they don't allow the app to be installed in my home country when I went there for a visit and was hoping to introduce my young nephews to the joys of learning a foreign language. It happens that third world countries don't generate enough ad money to be part of the free language learning for all mantra. Needless to say, I didn't renew my subscription. I must admit though that the app contributed toward my goal of attaining reasonable proficiency in Portuguese in one year, though I got much more out of obstinately listening to podcasts and regular radio even though I understood little at the beginning, and also making native speaker friends online. I used https://www.hellotalk.com/, an app to meet language learning partners to chat with, and where Brazilians happen to be extremely active and eager to exchange with people willing to help them with their English or French.
I use Duolingo for Portuguese and I would highly recommend sticking as much to desktop as possible, given that it requires you to work more at the exercises than on mobile. Better for ingraining concepts in your head.
Duolingo, along with other resources, gave me a great foundation for when I added Italki and started talking to Brazilians in Portuguese for 30-60 chunks of time. I probably should have started Italki ("out loud" conversations sooner), but Duolingo's Portuguese tree definitely gave me a great start on vocab and all the different grammar types (subjunctive in Portuguese is intense!).
Clozemaster has been my go-to app for language study now. It uses the "sentence mining" concept advocated by Glossika to show you thousands of sentences and learn new vocabulary in context.
After I got to a certain level, Duolingo and Rosetta Stone became too easy, and I found myself grinding on things that I already knew to try to get to "the good stuff". The signal-to-noise ratio was just too low.
Anki was the next step up for me and improved my vocabulary by leaps and bounds, but it became tedious to look for example sentences and make new flashcards from them. Clozemaster does this automatically, and you can choose whatever difficulty level you like.
That's the approach I've had recently. Duolingo into Clozemaster, with generous helpings of Memrise on the side. Too bad Clozemaster is awfully limited in its free version.
I'm using Duolingo to learn Hindi, but not alone; I also have a couple of (Snell's) books.
I find that a good combination - I favour reading as a means of study anyway, and that gives an understanding of the grammar, tips, gotchas, etc., and then Duolingo puts it into practice, tests me on it, and instills vocabulary and recognition of sounds and characters through repetition.
Since we're recommending other language tools, I've found the Michel Thomas method to be really effective. Better than Pimsleur: https://www.michelthomas.com/
Duolingo functions on the premise that "being conversational" means "cobbling together words shakily into passable sentences". I find that people in general have a low bar for "being fluent" and "being conversational".
Have you ever tracked a conversation you've had with your friends? A conversation can broach many different topics, and can go deep into those topics. Are you "conversational" if you can only say a few sentences and are completely out of your depth once the conversation switches to any other topic like politics, etc? No. Conversations can also last hours.
Doing things like duolingo will never have you reach fluency or even being fully conversational. It may be a good start, but it should always be complemented by studying with textbooks or online resources like instructors on youtube. Anki (or memrise) is necessary to start memorizing heaps of vocabulary words so you can begin to have simple conversations. Glossika is good for making yourself practice saying sentences and listening to sentences.
Native material is by far the most important aspect. That's how you find vocabulary and practice things like listening and shadowing along.
157 comments
[ 4.4 ms ] story [ 226 ms ] threadI’ve learnt languages with immersion and Pimsleur. Sadly I haven’t found Duolingo useful at all. The effort to actual useful knowledge ratio is high.
Other helpful resources: Assimil, Anki, and Conversation exchange. Mundo Lingo is good for that if they have an event locally.
I also did it in parallel with the Language Transfer podcast, which I've mentioned on HN before but is an amazing resource for understanding how a new language works structurally and grammatically.
I did find Duolingo valuable, but purely for vocab.
https://static.duolingo.com/s3/DuolingoReport_Final.pdf
A few notes about it though:
1. It was funded by Duolingo but carried out by an outside team of academics. I can't judge how much this impacted their analysis.
2. No control group, which isn't very promising for the rest of the methodology.
3. More time spent using Duolingo did seem to result in higher improvement
4. Lot of dropouts, some participants excluded for taking up courses. They tried to control for 'outside resources' like watching movies in Spanish but it's unclear if that problem was really solved.
5. Novices of the language learned the most. Makes sense, as Duolingo focuses a lot of vocabulary and basic sentence construction
Personally I'm using Duoling to learn Italian, and I find that I am definitely learning, especially in terms of conversational phrases and vocabulary.
However, I've learned a few other languages by immersion, and this is going much, much slower. I'm 99% sure I would learn more by moving in with an Italian family or by consuming all my media in Italian, instead of Duolingo. For now, Duolingo is a nice compromise, and I accept that I'll learn a limited amount from it.
Duolingo is a feel good tool - it makes you feel like your learning, perhaps quickly, without really learning anything.
For example, it uses multiple choice questions. Presented with a word (ananas), it then shows a few pictures (apples, bananas, pineapples and pears, for example). You incorrectly choose 'bananas', because really who wouldn't, and the pineapple lights up in green.
This method is really at odds with the 'best' ways to learn. If your recall is dependent on contextual cues, you will struggle away from that context.
Additionally, the volume really isn't that great. Like a lot of 'feel-good learning' platforms, duolingo encourages consistancy over volume. It's better to study for 10 minutes a day, than for 90 minutes every sunday. That's true, but if you want to learn a language (no small feat), then even 90 minutes a week is woefully inadequate. Being conversant in multiple languages is something to be proud of, and things to be proud of tend to take substantial effort.
Vocabulary acquisition is almost the perfect problem for spaced repetition, so use anki. There's a few strategies here: english word -> foreign word, picture -> foreign word, foreign word -> foreign synonym, foreign word -> english options, english sentence -> foreign sentence.
Pronunciation is difficult at first but gets easier. Accents are tricky - they're verbal gymnastics. Its easiest if you can find a native speaker of your target language, and you should just converse with them. Don't focus on minor details, because you usually pick these up naturally over time.
Listening is easy - listen to their radio, watch their TV. Similar with reading.
But there's one thing I haven't mentioned. Grammar. You do not need to learn foreign language grammar. You'll pick it up. How often have you studied English grammar? Unless you went out of your way, probably never. You learned some basic rules in your first few years of school, but everything else you picked up.
At least it saved me a hundred euros, but it's not a substitute for a real course and real interactions.
But later it talks helped in using more esoteric grammar that I never hear in small talk.
I'd suggest to use the website from time to time as it has some well done explanation to what is grammatically going on (the app is theory free)
I find that Duolingo is the least useful of these apps. Its main advantage is that it's easy and a small commitment if you decide to do one or two lessons every day. For the rest I don't understand why it's so popular. Duolingo's main strength is that it makes you actually construct the language instead of just memorizing individual words or expressions but I'm not super impressed with that either, mainly because in my experience it's very, very common to have a sentence rejected because it didn't match the internal "regexp" used to validate it even though it's perfectly correct (and sometimes these mistakes linger for literally years despite being reported). I actually got a very mediocre result in the placement test for the "French for English speaker" tree despite being a native French speaker who's reasonably fluent in English, mainly because some perfectly correct answers were rejected by the system. That really destroys your confidence when that happens (both in yourself and in the app).
You have a forum to discuss these issue which contains some very valuable information but it's the worst forum software I've ever used bar none, it feels like a teenager's first PHP project in the early 2000's.
Oh and if you use the mobile app it'll have you build the language by selecting one full word at a time, which means that you usually won't have to thing about the conjugations/declensions and just vaguely remember what word means what. It's fine for English, not so much for Russian and its complex declensions and aspect system.
For vocabulary Anki and Memrise easily win because they only do that and they do it well. You can find podcasts aimed at learners of many languages to improve oral comprehension and LingQ is great to improve reading comprehension, although it's expensive for what it is and I feel like you could make a better clone of it in one weekend.
In general I haven't been really impressed by any of these language learning apps, it seems that they really lack the resources to do anything but the bare minimum. It's probably too niche to generate some real R&D.
So overall if you enjoy Duolingo then stick with it, but keep in mind that you probably won't get anywhere just using this app. If I had to recommend only one language learning app it would be Memrise because it's got some decent decks for many languages (including user-contributed ones) although of course you can't learn a language solely by memorizing the dictionary.
* First and foremost, Duolingo isn't a product, it's a series of products, which all have different characteristics. A person using Duolingo on Android will have an experience completely different from the experience of someone who uses iOS. And the difference is even more drastic when comparing mobile with the desktop version. The desktop version is sometimes three or four time more challenging.
* Even within one platform, the A/B testing has become so large-scale that even two Android users might see a completely different product.
* Courses for “popular” languages get way more attention than the “unpopular” ones. And the quality varies greatly.
* The ads they show are sometimes loud and obnoxious, NSFW, or straight up scams. That doesn't happen that often, thankfully, but I've had my portion of loud-as-hell game ads and borderline pornographic hentai game ads. And before you say what people always say when it comes to NSFW ads, Duolingo says that ads aren't personalised, so no, my search history has nothing to do with them.
* Finally, it is indeed a good way to learn some basics and acquire some basic vocabulary, but there is no way you'll get fluent with it. A friend of mine has summed Duolingo up very well: it gives you a feeling that you learn something, even if you don't progress at all.
That's just the stuff I could remember off the top of my head. It's still nice to have it, but we still should remember that the service has lots of issues, some of which could be eliminated, if the management wanted to do so.
Also a bitch to maintain with three experiments that might or might not be going on at the same time in one screen.
It's pretty good for the first steps, and to get some vocabulary. I can't follow a non-basic conversation, but I can pick up words from native speakers.
I'm now at the point where I should work on my own with more seriousness.
Overall, it's good to start learning a language, and to keep a regular practice, but it's not enough on its own to actually speak the language.
I do 10 minutes per day, so it's fine by me, I usually waste more time in social media.
What I've realized is that experience teaches you to speak, read, and write a new language. You need a lot of experience to become good at it, because the language has to feel intuitive if you want to hold a conversation in it. There's the recommendation that if you want to learn a language you should immerse yourself in it (in real life). I think the reason why that works is because it forces you to figure out how to use the language and gives you an immense amount of experience in it.
I think that Duolingo is just another way to get slightly more experience in the language, but it's probably not going to be enough on its own. You're just not going to be spending hours every day on it to compete with language classes. From my experience, it was a nice way to quickly learn some of the basics of Japanese, but Duolingo really can't make you understand Kanji more easily than other methods.
I had no Spanish background before Duolingo, or any spoken language other than English. I took Latin in HS, which helps somewhat with vocab and reading but not at all with speaking and listening. With solely Duolingo plus the additional practices I mention, I've gotten to the level where I can survive in a Spanish speaking country. I can't speak or listen well, my grammar isn't good, but I know enough to ask for what I want and understand basic information being told to me. I can also read enough to understand at least the key information from most written sources. I think I'll definitely need a more dedicated class and/or a longer full immersion experience to get towards my goal of being conversationally fluent, but Duolingo has been a great starting point.
What I find pressingly missing is any meaningful way of learning about the grammar of a language. When learning German there was absolutely nothing in the lessons which even hinted at the rules of tenses, cases, conjugations etc and the lack of that content stunts the possibility of building much more than a rote-learning knowledge of a language.
The app does seem to have a great gamification mechanic and is good at driving user engagement, so I guess they have that though.
Click on a lesson's circle and there is now a lightbulb icon which opens the detailed lesson.
I’ve already learned similar languages the traditional way and I go to the netherlands at least 1-2x year for work.
Personally, I feel I get enough value from Duolingo to make it worth my time. But I admit I’m a particular use case.
IMHO it is not was way to learn language if it is one’s only tool. But then again neither is only classroom instruction or only books or only tv. I think that is where people get hung up on critiquing it.
I see it as ONE valuable tool in my toolbox.
And yes, I've dabbled enough to know that I could learn Brazillian Portuguese and try to convert, but I've also dabbled enough to know that to do so would be really hard, and leave behind some undesirable remnants.
No, that's not true. Learn Brazilian Portuguese and you'll be 95% there. The major differences are in the voçe vs. tu (second-person singular pronoun). There are other differences (mainly pronunciation), but the core of the language is the same.
Source: I'm from Argentina and learned Brazilian (travel often). I have a few Portuguese friends and I have 0 issues talking and understanding them.
Understanding and being understood is not the entire point here. Having said that, I'm finding virtually no resources for learning Euro-PT, so I'm likely to have to resign myself to always being thought of as a speaker of Br-PT, despite living a few hundred kms from Portugal.
(BTW - thanks for the response, it is a useful data point)
Nonetheless I suggested my mother and grandmother to start using it to get better at their english and they’ve been using it every single day since 8 months ago — and I’ve seen incredible results. They start to understand basic online content and they are asking me to translate content less and less every week.
This is an incredible result IMHO.
Are there any free and open source alternatives to duolingo out there? I wouldn't want to use something that could shutdown after an 'incredible journey'.
Anyway, I've been using Duolingo since ca. 2014 and it didn't shutdown... Plus, it's free - what do you have to lose? Your progress on some language learning app? You won't lose the knowledge gained.
I am quite curious how you figure this. Other than analytics and tracking, Duolingo doesn’t even require a backend.
Anki is useful for spaced repetition, but setting up a deck is going to take effort and requires knowledge (building a deck is useful by itself though).
Basically you get four hearts every few hours. Every mistake takes away a heart. If you lose all your hearts you can no longer play. But, you can do _review_ lessons (that have nothing to do with the content your currently learning) and regain 2 more hearts.
It basically became a grind situation and I stopped using it. I've heard the webapp doesn't use hearts, but I refuse to support them after that.
And, yes, I was planning to buy a program to learn Spanish and had been using duolingo for 2 weeks prior to getting the hearts.
I did some reading on reddit and it appears this is related to some VC funding and attempt to make money. I understand they need to make money, but this felt like a very bait and switch slap in the face kind of thing.
The whole experience kind of soured me on learning for now and I haven't been spending as much time on it since.
I did find clozemaster which seems to be a similar, and honestly better, app. I also have a textbook and workbook pair. And, fortunately, have a lot of native Spanish speakers around me.
I've messed around with Duolingo a few times, just to get a feel for how useful it is. I came away feeling that it was about the same level as flashcard vocabulary learning - useful, but only a small part of the pie.
Vocabulary is a necessary, but not sufficient, condition to be able to carry on a conversation. Mandarin in particular is more difficult, since you need to grind through hours and hour writing/rote learning characters to improve your vocabulary.
The one thing that noticeably accelerated my proficiency was having an actual conversation with a native speaker (host family/tutor/etc). That's not something that Duolingo really replicates.
Self-plug - coincidentally, earlier today I just released a small tool for Mandarin learners to practise their tone recognition! I've been working on a larger app and this seemed like something that would have helped me when I first started out. App is currently live @ https://fluent.academy - I'll be introducing more functionality for Mandarin learners in the near future, so would really love to hear from students of any level.
https://mandarintonetrainer.com/games
Swell idea though. Mobile support is a bit rough but not unusable.
Out of curiosity, what mobile issues are you encountering? This is actually a Flutter web app, which isn't officially production-ready yet, so I'm keen to provide feedback to the Flutter team if something's going wonky.
I've been learning French and have been using Duolingo for vocab, but then I listen to a French-teaching podcast as well as trying to work on reading/speaking it.
Regarding the differences between Canadian and France French, are there that many differences that it would make it difficult to understand someone from another region?
I have friends and former coworkers some from Quebec, some from New Brunswick all fluently bilingual. We've all worked with people from France on projects and my Acadian friends remarked they couldn't understand a word the people from France said.
Although the reverse of that a friend went to northern France on vacation. When he was there he said everyone asked him where he was from he told them New Brunswick Canada. From what I understand Acadian (French dialect of New Brunswick, Canada) is older and closer to northern France than the Quebec dialect. I'm not sure if my friend meant the regional dialect or standard Parisian French, I'm pretty sure he meant the former.
And this may be totally wrong but I believe the southern France dialect (Occitan) is quite different than the north. I've had to call Monaco (I know it's not France) and I was surprised at just the accent of the French person speaking English. I've been around French all my life but southern France to me sounds like stereotypical cartoon French like Pepe LePew.
The entire history of standard French for their language is quite interesting. It's not been accepted for as long as you may think and there are many dialects.
edit: I think Canadian French has a lot of slang not used in France which makes it hard to understand. Kind of like a Cockney from England coming to the US, sure you can understand his language but not the meaning the way he uses words.
With that said, it was reasonably fun, and I did stick with it for a while due to the gamification, so maybe there's hope for the feature. Just getting people to study a foreign language regularly is probably a good step.
I also think that they have implemented some more advanced learning features with a few languages.
How exactly were you using it? It speaks to you, which is invaluable for developing an ear for the language. It forces you to translate both ways, which helps with reading and writing immensely. And optionally you can even speak answers in the foreign language to practice pronunciation (such that your phone can correctly transcribe what you're saying).
It's not exactly an immersive language course, but it is vastly more useful than flash cards.
I know a lot of people like to speed through the tree and only complete the first level, which I suppose is a more vocab-focused way of using the app. But if you go deeper on each level it forces you to grapple with grammar a lot more.
Probably a fair point - as mentioned I wasn't really using it to learn the language, rather to get a feel for "how useful would this have been when I was learning". So I admit I didn't dive into the later levels.
2. italki for conversation with native speakers. You can find super cheap if you are just looking to talk.
3. hellotalk is free but takes more effort and mostly just spanish/english.
Asked me to sign in with Google. Couldn't click behind. Closed window.
Come on, don't make me sign in everywhere. This is just email harvesting. I get if you want to sell a service, but then I need to know what you are providing. If it is good I'll pay for it (as someone trying to learn Mandarin. There just aren't good tools out there). But please don't just blindly collect user data.
In the meantime, I've enabled e-mail registration without verification, so feel free to signup with a throwaway :)
Also, in counter part to Duolinguo, I suggest at checking out Hello Chinese, and using that as the metric to beat (it is popular among Mandarin learners).
You're right, there are a few artifacts in the audio. Most of these are due to generating speech at a slower cadence than "natural".
Native/enunciated recorded speech would probably be better for this specific purpose, but not necessarily practical for the larger app.
I'd love to try this but I don't have a Google account. Do you have any plans to add registration using an email address?
A couple of comments: for 水果, it only accepts shŭi gŭo, not shúi gŭo. I know that is technically correct, but the voice synthesis sounds it out with the sandhi. More a question of philosophy I guess, but thought it was worth mentioning.
Also, it might be nice to have an option that hides the characters, so you're forced to rely on hearing rather than just knowing that e.g. 多 is dūo.
Looking forward to seeing what else you do with it!
In general, I recommend Glossika or Pimsleur over Duolingo. Depending on the language you're learning, there are also more niche sites that do a far better job; Turkish Tea Time or RealPolish.pl are some good examples.
https://turkishteatime.com
https://realpolish.pl
I haven't looked at either of Glossika or Pimsleur, but Glossika's page title is "AI-based effective language learning" which immediately reeks of marketing blather. Why do you recommend them?
I’ve found this approach to be more fruitful than the abstract one most books and programs take.
Based on the screenshots, it looks like they maintain a "memory level" for each phrase and presumably do spaced repetition to keep that level above a certain target. Am I right in assuming that if you fail a review because you forgot what one specific word means, you'll get the exact same sentence next time, rather than a different sentence using the same word?
I have a system for personal use that does the latter, but I haven't yet found a good interface for indicating why I got the sentence wrong. My current implementation is to have a huge list of checkboxes for every aspect of each word (dictionary form, spelling, pronunciation, etc.) that could potentially be forgotten independently.
Correct. It’s just based on the sentence, not the words within the sentence.
I have two main sources of example sentences: 52114 from Tatoeba https://tatoeba.org , most of which have translations, and 114591 extracted from public-domain ebooks from Aozora Bunko https://www.aozora.gr.jp/ , some of which use rather archaic language.
I process each sentence using Kuromoji https://github.com/atilika/kuromoji to segment it into individual words and get details like part-of-speech tags, dictionary forms (lemmas) and pronunciation for each one. Then I add them to a SQLite database.
Some sentences on Tatoeba have audio recordings contributed by native speakers, but there aren't so many for Japanese, so I also use Open JTalk http://open-jtalk.sourceforge.net/ to get a robotic pronunciation. That pronunciation may differ from the one determined by Kuromoji, which I use as a consistency check.
I have tried various criteria for determining what to learn next. One that worked well is to keep track of the least frequent unknown detail of each sentence (e.g. the spelling of a particular word) and pick the sentence where that frequency is highest. That way, sentences with common words come first. Lately I've modified it a bit to also consider how many other sentences have the same least frequent unknown detail, so I always have multiple different examples available to help me generalize.
Reviews are scheduled for each detail based on the time since the last review and the time when I last had to relearn it. I've been collecting data to later tune the review schedule, but right now it just doubles the interval between reviews every time I remember correctly.
I used to just pick a random sentence for each detail with a scheduled review, but that would sometimes randomly schedule the same sentence often enough to be noticeable, so I've changed it to prefer sentences I've never seen or not seen in a long time.
The two kinds of review are dictation, where I listen to the recording and write it down on paper, and reading, where I read the sentence aloud. In each case, I manually check whether I made any mistakes and indicate them by unticking a checkbox (as mentioned in my previous comment). Those details will then be scheduled for review more frequently.
After a year of daily use, there are now 35611 sentences in my database which could potentially be scheduled for review (which means that, theoretically, I should be able to understand them), 21.3% of the total. The ebook which I should understand the best (35% of all sentences) is the Japanese translation of Karel Čapek's play "Rossum's Universal Robots" https://www.aozora.gr.jp/cards/001236/files/46345_23174.html 35% understanding is ... not enough to allow me to read it from beginning to end, but occasionally I get a nice sentence e.g. about the worldwide robot uprising.
I think it should be possible to adapt the system to work for other languages, but it would require significant refactoring to remove the assumption that everything is in Japanese.
One question that sticks out in particular: how do you actually perform the reviews? Did you create a custom GUI for it or are you using an off-the-shelf software?
I created the GUI in Qt. It's not beautiful, but gets the job done. I decided to finally open-source my WIP and put a screenshot in the README: https://github.com/Yorwba/alphabet-soup
The comment on database migration is here: https://github.com/Yorwba/alphabet-soup/blob/4f68057ed5d7dbf...
If you don't care about all the Japanese-specific stuff, you could try running make data/tatoeba_sentences_XXX.csv where XXX is the ISO 639 three-letter code of the language you're interested in. That collects all the sentences in that language contributed to Tatoeba by native speakers and puts them into a TSV file.
Although it's not perfect. I prefer Pimsluer over Micheal Thomas Method, which I've done previously. I'm just using it as a source of comprehensive input to develope my ability to able to understand spoken Mandarin (which MTM was awful for).
I'm sure I could use other programs similarly. Though Pimsluer probably has the most material. That might not matter if I get good enough to use films without subtitles and podcasts for my sources.
Glossika looks like it's geared more to how I'm trying to learn. I'll probably give it a try at some point.
I think that's the hard part about learning languages: it works way better if you don't do it alone, and that's why apps can be pretty limiting.
I have only given Rosetta Stone and Pimsleur a cursory attempt but they seem awfully slow and expensive for what they do. Duolingo, for all its flaws, is free and fun.
As for your "purely from marketing" comment - Duolingo solely relied on word of mouth for all of its early phase, apart from a single TEDtalk. It's true it may not be as good as many alternatives, but as far as the free and fun factor goes - it is simply unmatched.
To be fair. Rosetta Stone got it's "reputation" because of it's insane price point comparative to what you learnt. Duolingo on the other hand is completely free.
I do think that using duolingo too long is an inefficient way to learn a language even at beginner level, I wouldn't use it for more than a few weeks before graduating to a SRS like Anki.
For intermediate to advanced I really love clozemaster. It's especially good for learning grammar, you probably want to supplement it with some pure vocabulary in an SRS.
My experience was that I was much better at recalling vocabulary from using Anki than DL. Possibly because the "yes/no" to DL is at a phrase/sentence level rather than per-word.
Without putting effort in, DL makes it easy to recall what the DuoLingo-defined accepted translation of phrases is. It's easy to fall into the habit of 'gaming' it and just remembering the answer in a very shallow sense.
With Anki, using flashcards of words, the 'effort' you need to put in is being honest about whether you recalled the word.
I felt that when I started using Anki, I was much better at answering DL.
I find Duolingo to be an extremely fun game. My wife and I really enjoy trying to speak other languages and even more so now that we have talking children. I can speak passable Spanish and can string together broken sentences in French to be effective (they get what I mean) and hilarious to native speakers.
Duolingo has never purely taught me a language. If my wife didn't know Spanish and learning French like me, I would have never been able to practice enough in the real world to know.
I enjoy using Duolingo immensely. For me it is a fun way to toy around with a hobby I've had for years without committing myself to dedicated training costs and time.
Recently, I’ve tried writing smaller apps to supplement what Duo provides because they have a great product and I’ll never compete directly as a solo developer.
Anyway, any other small app ideas that people would like to see implemented? Game ideas that help with the necessary repetition, for example?
Also as someone trying to pick back up on my Japanese after letting it slide for a decade (well to be honest I never had more than enough to do greetings and basic restaurant ordering), many kana/kanji matching apps leave a lot to be desired. Katakana is a personal weak point and the apps I've tried all limit the extent of matches to picking one of the matching 3 or 4. Process of elimination makes it so that I can get 100% of these quizzes despite only knowing 80% of the material. Something that would allow customizing 10 or 20 possible answers to pick from would be nice. But I don't know if this would have any bearing on other languages to be worth developing (and there might be an existing app I just haven't found yet).
That idea doesn't scale. Much simpler to look at definitions and example sentences, especially for non-obvious words like abstract ideas, context-dependent words, and particles.
> Katakana is a personal weak point
Just use flash cards. It's rote memorization, so there's no substitute for time and repetition. If you're able to commit an hour each day, you should be done by the end of the week.
>Just use flash cards.
That is effectively what the apps are mimicking and what I am using. What I'm bringing up in a desired feature I haven't found (granted, I've only checked out a dozen or so apps which is worlds apart from serious market research). As for why I use an app over flash cards? Phone is more convenient and always on me. Also, other than a stack of the most common kanji, flash cards don't really scale in a portable format.
But I probably should buy a deck including kanji for at home usage.
And I'm only recommending flash cards and other rote tools for kana, because recalling them instantly is both important and relatively low-effort. I don't have a strong opinion on how to learn vocabulary, other than to prioritize based on actual usage: media you read/listen to, conversations, etc.
They're also each a dollar to unlock voiced/combined readings and remove ads. My pihole did block the ads while I was just trying them out.
My exact experience. Been wanting to learn Dutch for years but I just can't invest the time and effort. I've got a 300 day streak on Duolingo now and I'm close to A2 level and I can already participate in some online conversations and understand a great deal. It is language learning for "free", I'm not talking about monetary cost here. I wanted to learn a new language but didn't want to put the effort in. It feels like I'm learning for free, without the effort which is exactly what my lazy ass wanted.
If I were in a hurry, or wanted to put in serious effort, it probably wouldn't be the most efficient way. But I'm not in a rush. English is my second language and my formal training brought me to B1 level at most. The rest I learned through exposure online over the years (and got an almost perfect TOEFL score a decade ago) - I assume the same will happen with Dutch too - I got a foot in the door thanks to Duolingo. I don't care if it takes a couple more years compared to another method that I'd probably never consistently use.
I do have tons of twilio/glitch based reminders that I moved from crons on a VPS in 2017. I am aided by technology to remember, but I usually do it early in the morning.
What I like about Duolingo is how casually it approaches you to make consistent efforts every day, without overexerting you to train, like many fitness apps usually do.
So far, I'm quite happy with it.
> I can speak passable Spanish and can string together broken sentences in French to be effective
Things like this worry me. If we look at how long Spanish and French take, it is about 600hrs[0] (probably less because specifically French is really close to English and Spanish is close to French). If you only had 20 minutes a day you should be reasonably fluent. But if your wife and you are talking, I'm guessing you average more than that.
[0] https://qph.fs.quoracdn.net/main-qimg-688ca4d9dd1f88f5f655be...
This is a game that I enjoy playing. I have never tracked how long I take. Many days, I do only a single lesson. That's less than 5 minutes, sometimes less than 3.
I'm not trying to be fluent. I'm not really _trying_ to achieve anything. The closest I have to a goal is that I enjoy having another language's vocabulary that I can use when speaking around my children.
What's a bit distubring is that it is being taunted as a complete system for beginners.
I think it's a disservice to new solo learners since eventually people give up and have not learned as much as they could have given the time and effort put forward.
I think Duolingo has a place but it should be an aside to a proven learning system not the only tool.
I recently realized is only valid if you don't want to put the work in. People who are not willing to put serious effort into learning a new language will find Duolingo useful.
It's also useful if you come from similar language. E.g. you speak Italian and you want to learn Spanish (latin langauge). Or you're Czech and you want to learn Polish (slavic language, same grammar and compsition)
But if you are serious about learning a language, Duolingo will only slow you down. You learn conversational stuff and progress often with big holes in the knowledge and understanding the core principles and structures of the language.
For serious learners of a new language, keen to progress properly and fast Duolingo should be avoided.
Take English for example, difference between past simple vs. past continuous is easier explained once than randomly coming across on Duolingo never fully comprehending the principle. Therefore it's faster.
And form someone coming from non-Germanic language this will take time to comprehend, kind of like when you first learned some math concepts as a child, some took a while to click.
Now take words and their past forms, give, gave, given.
In order to use it in a natural way, you first have to drill and memorise many words ... not possible on Duolingo as one lesson you work with a bunch of "gave" and suddenly several lessons you use "given" ...
Yes, you can learn language in way that you go China and you're able to make primitive conversations. I know a lot of people speaking French that they learned conversational way and all of them struggle with grammar and despite being able to talk and communicate fluently their language is not enough for professional level or academic level.
E.g. my girlfriend and I speak English, a language native to neither of us ... we can live a complete life together and we can talk but to a native speaker her mixing of tenses and grammar would sound like nonsense.
This is the case with many of my foreign friends who learned English in a conversational way.
Duolingo does the best what it's designed for. It's a way to learn for the lazy, so it is a game.
You also conveniently ignore the countless people (including adults) who reach fluency without even studying, instead just picking it up from immersion: work, media, etc.
Duolingo is a specific tool. Like any tool, it's best for specific uses. Your judging its users as "lazy" is both baseless and unhelpful
Immersion is still studying. Digesting actual native material ranges from hard to very difficult (i.e. dramas vs comedy shows, etc). Immersion + studying textbooks is 1000x better than gamified sources like duolingo, particularly for Chinese/Korean/Japanese.
Also without knowing your native language and target language, the comparison is not useful. Spanish is quite easy for English speakers; good luck learning Cantonese without studying [for a very long time].
Nope. Not everyone has a lot of time to dedicate to studying a language, which is fine. But duolingo is never going to give you any meaningful fluency. It's like saying "I want to get ripped by only doing body weight exercises for 5 mins/day". If your goal is just to say "Hi my name is X, where is the bathroom", then by all means use duolingo.
> nobody can ever become fluent in a language from a class
No class will ever teach you everything about a language, partly because there is way too much to teach, and partly because languages change all the time and classes can't keep up. Instead, classes give you a starting point to start reaching fluency, which begins once you start consuming native material and having conversations with real native speakers. Classes don't teach you slang, pop culture, casual expressions, and so on.
Fluency is not making simple paragraphs or having a very basic, limited in scope & topic 10 min conversation. Language learning literature and other resources have lowered the bar for what "fluency" is, because nobody can sell a book or other resource by saying "use this book and learn french in 3-5 years". There is a lot of snake oil in this area, but people also want to believe in this notion that they can fluently speak a language with minimal effort.
Yes people say they used Duolingo to being close to fluent, most people who speak very bad grammar and mix tenses but people don't correct them mush as they get the point. So most are unaware. I speak it from first hand experience knowing about 10 people like that.
I disagree with your analysis. Most people aren't trying to be speaking super gramatically correct sentences. Most are learning to speak passable sentences.
Most important thing while learning a language is to have fun, so you can keep coming back to it. Duolingo makes it fun and interactive. Doing endless grammer textbook excerises /drills takes enormous will power that casual learners don't have.
Also a lot of people speak here from English native point of view ... yeah it doesn't matter you don't speak proper grammar when in Spain once in a while or while nomading. It does matter when you're foreigner and want to build a career or be successful in business/society in the said country.
If you actually want to learn a language?
- find one or three good textbooks and follow them completely with exercises
- find a language exchange, but it won't be helpful for speaking until you know 200-300 words and can speak a little. exchanges can be good for making friends, though. they're great for getting feedback on your pronunciation, speaking, etc, but always take things with a grain of salt. native speakers are not always good teachers.
- find graded readers or native content that is catered to early, medium, advanced levels, etc. immersion is really important
- use Anki for spaced repetition, to memorize vocabulary (that you found from native content like tv shows / books / etc)
- make friends thru language exchanges or apps like Hello Talk / Tandem
- in general try to get as much input and output as possible. people usually neglect speaking and listening.
- finding a weekly tutor can help you iron out personal issues, particularly with pronunciation
Doing an app for a few mins a day is never going to teach you a language to a "fluent" degree, despite what marketing they have. Learning a language requires serious study and practice- speaking (and writing, etc) are muscles that you have to train over time by interacting with real people.
I don't consider making a few contrived sentences like "my name is X, i do y for a living, where is the bathroom" to be fluent. Not at all. That's called being barely functional. Real fluency depends on your native language and target language, but requires years and years of work. It's not sexy like the idea of using an app and magically becoming "fluent" in a language.
Spanish with Duolingo, was at least pleasant. If I watch a series like La Casa de Papel or Orange Is The New Black I understand some words (more than something in French).
How would it have been vice versa for me? I will never know.
It's easy enough to get going and hard enough to feel like you're doing something but not so much you give up.
The points system, scoreboard, and email prompts are a good way to encourage you without being annoying. The ranked scoreboard really has me wanting to stay there and compete to stay in the top 10 each week.
One thing I miss is the speaking option. Not all languages offer it but some did and that option seems to be missing or at least on my device it is.
Another problem is some languages are hit and miss with pronunciations if there are few native speakers. I know there are computer generated voices but sometimes there is no sample of the words at all (Irish being one example).
Anything that gets people interested in languages is a good thing. Duolingo may not be perfect but it's a popular and easy to use app. We unilingual English speakers need to get to with the times and learn at least one second language, maybe even more.
> Anything that gets people interested in languages is a good thing.
I don’t share this view. If you want to really commit to learning a language for some reason, that is fine and great and there are now countless resources to do so, getting language partners is easier than ever... but it’s not a moral imperative, and if you’re just spending time on activities that have no hope of leading to fluency it really is just a mastubatory exercise (albeit one that is more appropriate to talk about in mixed company), or basically a game... which there is also nothing wrong with!! I masturbate.. but then again I don’t lord it over everyone as some moral imperative.
The more serious point: if Duolingo gets your rocks off, great, but as language learning tool it’s pretty poor. Sure you have to work to learn a language, but if you’re going to commit to that effort there are staggeringly better options, which given all the hype is a bit disappointing.
> Anything that gets people interested in languages is a good thing.
Whatever. If you want to really commit to learning a language for some reason, that is fine and great and there are now countless resources to do so, getting language partners is easier than ever... but it’s not a moral imperative, and if you’re just spending time on activities that have no hope of leading to fluency it really is just a mastubatory exercise (albeit one that is more appropriate to talk about in mixed company), or basically a game... which there is also nothing wrong with!! I masturbate.. but then again I don’t lord it over everyone as some moral imperative.
I would have been prepared to pay but I don't see why I would. The only paid features I see are to remove ads (which you rarely see compared to how long you use the app for), offline use (which isn't an issue if you have good mobile internet) and to skip to harder stages (which seems counterproductive).
In terms of making an income, it feels they give far too much away for free. Do they make most of their money from the ads? Or so many people use it for free that it's okay if only a low percent convert?
Duolingo, along with other resources, gave me a great foundation for when I added Italki and started talking to Brazilians in Portuguese for 30-60 chunks of time. I probably should have started Italki ("out loud" conversations sooner), but Duolingo's Portuguese tree definitely gave me a great start on vocab and all the different grammar types (subjunctive in Portuguese is intense!).
After I got to a certain level, Duolingo and Rosetta Stone became too easy, and I found myself grinding on things that I already knew to try to get to "the good stuff". The signal-to-noise ratio was just too low.
Anki was the next step up for me and improved my vocabulary by leaps and bounds, but it became tedious to look for example sentences and make new flashcards from them. Clozemaster does this automatically, and you can choose whatever difficulty level you like.
I find that a good combination - I favour reading as a means of study anyway, and that gives an understanding of the grammar, tips, gotchas, etc., and then Duolingo puts it into practice, tests me on it, and instills vocabulary and recognition of sounds and characters through repetition.
Have you ever tracked a conversation you've had with your friends? A conversation can broach many different topics, and can go deep into those topics. Are you "conversational" if you can only say a few sentences and are completely out of your depth once the conversation switches to any other topic like politics, etc? No. Conversations can also last hours.
Doing things like duolingo will never have you reach fluency or even being fully conversational. It may be a good start, but it should always be complemented by studying with textbooks or online resources like instructors on youtube. Anki (or memrise) is necessary to start memorizing heaps of vocabulary words so you can begin to have simple conversations. Glossika is good for making yourself practice saying sentences and listening to sentences.
Native material is by far the most important aspect. That's how you find vocabulary and practice things like listening and shadowing along.