Depends Leon. In US where there are Indian immigrants some places have mainly Telegu speakers, not Hindi. Not sure about Tamil.
Frankly with Catalan being up I’m a little surprised Basque isn’t up either.
My guess is this isn’t going to come out anytime soon. If Vietnamese doesn’t come out then I unfortunately have no reason to renew my membership next month. ![]()
I also don’t understand why there isn’t more uptake in most apps for Hindi (Duo and 1-2 others). Hindi is the 3rd most spoken language in the world, so it’s baffling (to put it nicely) to me that 30th-40th rank languages show up in more learning apps. Plus 1/3 to 1/2 of other-than-Hindi/Urdu Indian language speakers have Hindi as an L2 or L3, even more so than English.
It is definitely very frustrating.
I suppose most language learning programs don’t believe the financial rewards are worth the effort. Pity!
Please add Vietnamese even without the ministories in order to be able to upload at least our own content, books, articles. Thanks
I sent an email to Lingq support around 2 weeks ago. They said that they are planning to have Vietnamese added in early 2024. Very excited to be able to use the platform to study Vietnamese ![]()
That’s great to hear. Hopefully Hindi will be released at the same time.
Its great hearing that Vietnamese could be realeased soon. As someone with imigrant parents i never actually got the chance to truly dive into the Vietnamese language since the only ones im talking Vietnamese to are my parents and relatives, which I only encounter once or twice a year. Since im only communicating with them in simple language about simple topics, I dont really have a lot of vocabs nor the grammar or certain set phrases to express myself. There arent many people as well which want to learn vietnamese in comparison to chinese or japanese, which makes it even even harder to find qualitative material.
Thats why Im happy to hear that It could be released very soon.
Btw. happy new year guys!!!
Hidndi has just released.
And Vietnamese is still not there ![]()
If Vietnamese is planned for early this year does that mean someone has recorded all the necessary stories that were missing? If there’s anything that still needs doing to speed up getting it added then it would be good to know. I, and I’m sure many others, have been looking forward to using lingq for Vietnamese for some time now.
Biiiiiiiig
The time has come! Lets go!
Nice!
I’m wondering why wasn’t this advertised earlier, for example in one of Steve’s videos.
Looks like the milestones for ‘known words’ will be:
680 Beginner 1
2040 Beginner 2
8160 Intermediate 1
16320 Intermediate 2
28220 Advanced 1
41140 Advanced 2
Not sure how it will be possible to reach the higher intermediate and beyond since most of the words are in pairs of two’s but they treat words individually
And Khmer has the opposite problem, they teat whole sentences as one word. I hope they fix this
As mentioned, the beta Vietnamese does not split words properly. There are algorithms for Vietnamese word-segmentation, such as [1906.07662] State-of-the-Art Vietnamese Word Segmentation, but I tried out a very simple approach that recursively searches a dictionary. To illustrate, look at this sentence (I added the to properly segment words:
Vinh đến [cửa hàng] [tạp hoá] để mua [đồ ăn].
Vinh goes to the grocery store to buy food.
[cửa hàng] [tạp hoá] = store + grocery (correct)
cửa hàng tạp hoá = door + rank + miscellaneous + chemistry (LingQ)
To achieve the correct segmentation in this case: A simple algorithm grabs 3 words; i.e. “Vinh đến cửa”, not found in dictionary, so moves to 2 words; i.e. “Vinh đến”, and finally “Vinh”, next “đến cửa hàng”, not found, etc. for “cửa hàng tạp”, then “cửa hàng” that word is found in the dictionary, so group [cửa hàng]; this works quite well and it is pretty simple implementation.
Another issue is that “Vinh”, which here is a name is rendered as “honor”, even with a decent vocabulary, you might recognize “Vinh” as a person’s name or city in Vietnam, but not as the word “honor” (vinh dụ). So, should you mark as “learned”, since you know it in the most common ways, or should you mark as to be learned, but not in the context of the text?
Finally, for now, I create my own compound words, the system seems to recognize these in future texts and also the translations hold, but the problem is that when I mark “cửa hàng” as one word meaning “store”, I have to commit on “cửa” and “hàng” as well, even though I don’t want to learn “hàng” in this case, for example, and “cửa” does not really mean anything here.
For Mandarin, LingQ 's segmentation works quite well, Vietnamese seems very similar to me, so should be able to work similarly in the future.
@tgredig appreciate the feedback. We will be working on improving the word segmentation and translations.