Lapis app: a segmentation engine that understands grammar, the best dictionary lookups, SRS designed for language learning

This might be a problem from what gives us the current time zones and daylight saving times, I’ll double check. The reason we require this is that a mismatch will probably produce wrong daily schedules (since the schedule for the day has to know when your day starts and ends). I’ll confirm when I check.

Looks bloody good. I think it will work for a good place to feed in random vocab as you come across it and want to SRS it.


Should be fixed now. This is now processed in the background (dealing with timezones was not accurate enough when it comes to daylight saving, depending on the server host). Let me know if this works for you now.

1 Like

Perfect! Thank you!

1 Like

Hey everyone.
Thank you to the 300+ who have tried Lapis, and especially to those who offered feedback and impressions in this thread and in other places!

We started this with a Patreon-based financial plan, but over the last few days we faced some problems with that (if you’re in an unlucky region you can face some blocking issues). Patreon is now out of the question. Later, we’ll be moving to a subscription based plan (details TBD), but for now the app will remain free for as long as we can in the beta phase.

That said, we were able to setup a donation option for those who expressed interest in wanting to support the project. You can find it in the landing page ( if you scroll down. Your help will be reflected directly on how much time and effort we can spend on evolving the app to its original target goals and beyond. Thank you!


Congrats on 300+ users, that’s a big achievement! You guys should be proud!

I believe this is better for both you guys and the app in the long run! Looking forward to seeing what the future will bring :smile_cat:


I’m really enjoying the app, especially the grammar construction box! It seems a lot more accurate after the updates too :slight_smile:
I have some small notes (I didn’t take screenshots, sorry!):

  • 今日は displays the definition for こんにちは, though I’m fairly sure きょうは would usually be more correct
  • ケンカ doesn’t display the definition for 喧嘩, although I think it’s commonly written in katakana.
1 Like

Thanks! Both fixed.

Fixed a bug for when an entry has a valid kana only (usually katakana) reading. So other similar issues to this are fixed too.

1 Like

Just one piece of feedback, is a username really necessary?


When I saw this field, it stopped me in my tracks and made me deliberate potential usernames and what the use cases would be when really, the sign up form should be as simple to fill out as possible . Lapis doesn’t seem like a very social platform so surely having my email is sufficient?


Hey! A couple more notes:

  • おれ seems to default to “to be” instead of “I”
  • The segmentation feature doesn’t seem to recognize the imperative form of verbs: for ichidan verbs nothing comes up, and for godan verbs, it gives the construction for the -i form of the potential verb

Also, suggestion: would it be possible to include the “drill into this box” function on the SRS cards? Right now that option is only available in Workspace, but it would be helpful in Srs too if I wanted to save a box for it’s grammar.

It’ll become necessary when we introduce vocab sheets (you’ll be able to share them and allow read/write access, and maybe more. Great for book clubs). There are a couple more things we’re thinking about that might need usernames later, so we require them from now. We can’t use emails because then your email would be exposed.

“Name” on the other hand is completely useless, it was there because it’s included by default in our identity provider. We’ll remove it later.

In the backlog. One thing to note is that we’re not adding things in priority of commonness because we have a certain process of registering new grammar. That’s why there’s a couple simpler common forms not recognized yet, we’ll be getting to these soon.

A reason for that is we take a minimal snapshot of what you selected, whereas the data you see in the box is very rich so we didn’t include it in the snapshot. (It’s a snapshot and not a reference because the segmentation engine is changing all the time, so there is no valid identity in the sentence to track).

Remember that you can always click on “Segment this sentence” to view the whole sentence segmented again:

From there you get the full featured segmentation page again without leaving your review and you can drill into boxes. Would this be good enough for what you want?


Oh, I didn’t realize it did that! Thanks, that works :slight_smile: