AuthorI'm Gabriel Wyner, author of Fluent Forever. I'm also running a Kickstarter for a language learning tool that's become the most funded app of all time. AMA!
Oct 3rd 2017 by gwyner • 39 Questions • 106 Points
My name is Gabriel Wyner. I'm a former opera singer with a mechanical engineering background, and in 2014, I wrote the book Fluent Forever: How to Learn Any Language and Never Forget It, where I look into the science of memory to find ways to reach fluency in foreign languages faster.
2 weeks ago, I launched a Kickstarter for a language learning tool that's become the most funded app of all time.
AMA about languages, memory, opera singing, living in Europe, dog training, treadmill desks, whatever. I'll answer juuust about anything.
Edit: This was awesome! Thanks so much everyone! If you have more questions, keep dropping them here and I'll answer stuff as it comes in.
Proof Pic: https://i.imgur.com/S8I0FWB.jpg
Doggo Pic: https://i.imgur.com/k0r23kJ.jpg
seeing your rewards, how do you justify $200,000 to build a desktop app of a flash card program?
So when I started this project, I had around 70 development houses I was looking through, which I weeded down to a 5-company short list. We had a bunch of talks and collected a bunch of estimates. Those estimates blew me away. They ranged from $175k-600k. I ultimately hired two of them for Proof of Concept work, just to see what the experience was going to be like working with them. The best of the companies was REALLY good, and they were, unfortunately, REALLY expensive.
Just to add a native iPad/Android tablet app (which I'd imagined would be not PARTICULARLY hard) was an additional $100-150k.
I guesstimated my stretch goals, but I think in light of the iPad port estimates, I don't think $200k is unreasonable at all.
That said, there's a whole different discussion as to whether I should be hiring such an expensive company in the first place, when I can get it done much cheaper through hiring my own programming team or just hiring a cheaper dev house. In my opinion, this whole app hinges upon quality execution. There are other flashcard apps that already exist. My primary goal is to make this thing phenomenally good. So I'm going to do whatever is needed to hire that company. They really were that good. :P
I don’t consider English all that inflected...it only has around 5 verb forms, 1 adjective form. But maybe I’m not thinking of something or your sense of ‘inflection’ is broader than mine. What’s your sense of it?
What lead to Spanish, French, German, Italian, Russian, and Korean to be the first languages you're supporting?
(I backed mostly because of French and Korean...just curious how you got to these languages)
Cost + Interest.
Hebrew + Arabic require additional development of the UI to handle right-to-left stuff, so they needed to be in their own stretch goal. Dutch and Portuguese weren't very popular amongst the pronunciation trainers. So I lumped those into the same stretch goal.
Mandarin and Japanese are their own special dev/design nightmare, and definitely needed their own stretch goal to accomplish.
So the remaining languages we already had trainers for were Spanish, French, German, Italian, Russian, and Korean
Hi, Gabe! I'm one of your Hyperglot backers, and I'm looking forward to chatting with you! I'm also a (former) singer, and I think that's like 30% of why i decided to back you :P
Do you have any advice for people who are probably great at languages, but can't retain vocabulary? I can break down the conjugation tables of French and German right now, but I can barely ask my Oma about her day. I feel like I could get further if I could just retain more words.
But overall, I think to the extent that you focus your studies on learning words in the context of actual stories, those words will stick better.
Would gif support allow sign languages to be "unsupported languages"?
The problem with sign languages is the lack of a written language. This thing lives on written stuff. I feel like a sign language trainer requires some real re-thinking of this whole thing.
Loved your method for becoming fluent in French, it's the real deal!!! planning on using the smartphone app for learning spanish. What date should I be telling my friends for the beta release for French/Spanish?
I'm aiming for a February release of the beta for Spanish. Potentially a March release for French.
in the breakdown of the rewards you write things like "enough app usage for fluency in a level 1,2 or 3 level language. How are you defining fluency? that's the ability to order a pizza/ taxi or actually sit down and be able to speak for hours with people on a broad range of topics? also you write that based on your experiences 30 - 45 mins learning a day can take you to fluency in a level 1 language, for example, in 6 months; was that time spent purely reviewing cards or that included the creation of cards? if the latter then presumably, given the often tedious amount of work involved in creating cards - which hopefully this app is going to completely remove, that time would be much lower - like 15 - 30 mins a day? i presume that one is going to be able to copy in text and have translations and pics supplied through the various resources the app is going to be drawing on? that it's not just going to be available for the vocab lists provided? it's most probable that stretch goals like the addition of monolingual dictionaries aren't going to be reached (though we live in hope). In this event how likely is it that you're going to be able to get these missed stretch goals in asap? (are they really going to cost that much more for each addition? that's crazy). also how data hungry is this likely to be? ie if i'm not connected to wifi but rather on mobile data is it going to use up loads of data with all the searches and downloading content like pictures? how is the offline mode going to work? firstly is that included as part of the main product and not a stretch goal? secondly, how much is it going to cache? would i, for example, be able to cache enough material for a 45 min train journey with no internet? is there going to be a function whereby i can just copy in lots of text (like with supermemo's incremental reading) which i can then come back to later (would this also have translations, pics etc cached for offline mode?)? is going to be possible to have these blocks of text separate onto different cards automatically - ie if i copy in a block of text that is 10 sentences long then can it automatically create 10 separate sentences that will be sitting in a queue waiting for me to get to them? i think that's about everything for now :-) Thanks for this! big fan of fluent forever. it has been a massive help and i can't wait to see the process become exceptionally streamlined and portable. i've donated and spread the word. lets hope it picks up massively!! :-D
You're asking a bunch so let me know what I miss:
How am I defining Fluency: - I'm aiming for around B2 or C1. If you're doing that lower end of 30 min/day, then more like B2.
What's that time spent on? - Both card creation and review. And I'd stick to that 30-45 min amount. Worst case, you get too good at your language. :)
Missed stretch goals: - We'll add them as soon as we can (so I guess that's "ASAP"), but I'm not sure when ASAP actually will be. I certainly want them in pretty rapidly for my own studies, so I'll be pushing for going fast.
How data hungry? - Not sure! I don't see why we need to be downloading super high res images, and mp3s aren't HUGE, so...medium data hungry? We won't really know until we get in there and do it.
Offline mode - Aiming to precache searches for the sentences that a user is expecting to play with. We might be able to predict how rapidly you eat up sentences and recommend a quantity for you, based upon your trip length.
Large texts - Yeah, we want to have a way to bring in a bunch of text, break it up into sentences, and have them in a queue.
OK I think that's it! Let me know if I missed anything!
First of all, thanks for putting so much effort into making all our lives easier!
For the app, will you recycle the audio files you created for the 625 word lists or will you re-record them? I am currently using the Korean one and realised there are a few files that are cut off or where the speaker sounds somewhat unnatural compared to sound bites on Forvo (this one might just be my untrained ear though).
Also, will it be possible to choose between different sound files - same as is your plan for images - and will I be able to add my own?
I'm not yet entirely sure how much we're going to re-use single word recordings, rather than simply having full sentence recordings for everything.
We may well reuse the 625 recordings, so if there's stuff that's meh, email us so we can flag that stuff for re-recording.