Live data from Hacker News

Show HN: Phrasing – learn every language, to any level

phrasing.app

61–70 of 126 posts

Re: Show HN: Phrasing – learn every language, to any level

#61
post #47

Given the prominent mentions of OpenAI I'm going to guess this doesn't use actual human translators but instead relies on machine translations using OpenAI? This is literally what created a drastic decrease in the overall quality of Duolingo. I wouldn't trust an app like this if it's relying on machine translations, AI or not. Especially when it comes to learning phrases and real-world examples. Translation is hard.…

Ideally we're not creating any of the translations ourselves - although of course we support machine translation. Our goal is to turn any native (or professionally translated) material into language learning materials.

We also have lots of ideas for 'Humans in the loop', verification, etc. But while we use AI for _a lot_ of things, we're trying to stay as far away from 'generate some Sanskrit' or 'translate to Sanksrit'

Re: Show HN: Phrasing – learn every language, to any level

#62

The page doesn't make a good job communicating what it offers. There's a lot of unimportant fluff, generic "we're the best", but I didn't really understand what the app offers beyond showing the same phrase in multiple languages.

Try https://phrasing.app/

The /playground link I posted to HN was meant to give HN something to click around on. The root index page is meant to be the primary marketing materials.

Re: Show HN: Phrasing – learn every language, to any level

#64

Earlier quoted context omitted.

Why? Everyone acquires language the same way.

Their first one, maybe. But I've seen first-hand how much people vary in what they manage to learn from classes.

Well, there's not that much variation. People don't learn much from classes. And you'll note that "what works for polyglots" isn't classes.

Re: Show HN: Phrasing – learn every language, to any level

#65
So what does it actually do to youtube videos? The landing page doesn't cover this. Should I expect that the video I watch will be in my target language now? As in, it'll be dubbed in the new language? That'd be super useful but I'm unclear and I'm not paying $10 to find out.

A time-limited example or something similar would be really useful. Maybe just one that you pick. Whatever it may be, I'd love to see how it actually works before paying money for it. Assuming it dubs videos, how does it handle multiple speakers?

Re: Show HN: Phrasing – learn every language, to any level

#66
post #56
post #53

The French example is horrible. Barely sensical, very weird, absolutely not what a native person would write nor say. It looks like something someone who barely speaks French would produce.

Agree. As a native speaker, I found these examples to be weird at best. No way any native would write or speak like that.

When we built this demo we figured the official translations would be well translated, apparently they are not.

Some people elsewhere in the comments are saying it's just a weird English sentence so it has a weird French sentence. My French is no good enough to critique on it's nativeness though

Re: Show HN: Phrasing – learn every language, to any level

#67
post #27

I look forward to giving user feedback when I’m able to signup. > our Humane Spaced repeition system. Based off of the popular FSRS, our Spaced Repeition System will adapt to your review style. Take a few weeks off? No problem. Want to study ahead one day? No problem. Took a 2 year vacation? We're jealous, but we've got you covered. I really resonate with this. I find that after a long break, I have forgotten a lot l…

I know Anki uses a modified version of SM2 so I can't say for certain, but if I recall correctly they are still using a linear algorithm. Most forgetting curves are, well, curved. More logarithmic.

Once you go too far away from the 'due date', the algorithms decent quickly to "0% chance of recall", where in actual memory it's more similar to a half-life.

It's been a couple years though since I've worked with that algorithm

Re: Show HN: Phrasing – learn every language, to any level

#68

> The hardest part about learning a language is knowing what to learn. I disagree with this. The hardest part about learning a language is finding content and resources in the target language that are both engaging and comprehensible at your current level. It's hard to find books, shows, or conversation partners, all the way through that critical awkward stage from beginner to vaguely proficient. The parts of the lan…

Nicely said. It seems like you're familiar with Krashen's Comprehensible Input Hypothesis? My daughter speaks German and Swedish very proficiently just from watching hours and hours over a few years of ZDF/ARD/usw (German tv networks) and Barnakalan (Swedish kids network) on the iPad. I can't say it was a wise to let her be on the iPad for that many hours a day, but it did work, she did acquire the languages.

Re: Show HN: Phrasing – learn every language, to any level

#69
post #15

Earlier quoted context omitted.

I believe that came from the official translation for paul graham's essay, but their site is down now and I can't confirm: https://penelope.jdsfriends.com/DWYL_francais.htm Not saying it's correct, but it's definitely only going to be as good as it's inputs. If we gave it a bad translation, that's on us! Sorry

If you do any sort of automatic attempts to discover parallel texts (and I can't understand how you can support "any show" otherwise), you're going to run into the problem of terrible translations, most of them made with machines.

You're also going to get figurative translations instead of literal ones.

Re: Show HN: Phrasing – learn every language, to any level

#70
I looked at some of the Hungarian translations and some are way off or even unrelated to the English version. I would have thought a year or two ago that this is simply a database lookup / array indexing problem, or is this something that qualifies as an LLM hallucinating? The former issue could be confidently resolved in a matter of hours, the latter makes me completely distrust the whole project.

I do like the UX tho, and find it a good idea that you created a HN-related landing page. Seems like a good amount of work went into this, congrats!

Post reply on HN