@mike I rather had that idea (see the pic) since it looks more natural to me, but if it can’t be this way, I think it will be okay as well.
As for the SRS — yes, I like it! As of now, there is no SRS, right? Because when I did my review sessions, I noticed the same sentences appearing again and again. Could SRS work like in Clozemaster?
Thanks for the screenshot - that’s helpful. When you click the Review button, should you go to existing review screen where you only see one sentence at a time, or should you stay on the article but have the text hidden for all completed sentences?
And thanks for the input on the SRS. Correct - as of now there’s no SRS, the sentences are simply selected at random. Yep I’m thinking either SRS like in Clozemaster. I know a number of users on Clozemaster have also requested a SRS closer to Anki’s algorithm, so may try that here as well.
I don’t think it makes a lot of difference… I personally like the first one since this way you clearly see that you are doing a review, and it’s also similar to Clozemaster, which I believe TG2 users are used to, but maybe create a poll to ask what everybody thinks?
Would it be possible to add a tureng link within this popup?
For example, for this word “olay”, it would be:
https://tureng.com/en/turkish-english/olay
@Punk added!
@Kaspac updated here New features: Phrase Deck, article reviews, and URL shortcut. Added the review options via the dashboard to save clicking into the article to review. Interested to hear what you think! Will also keep you posted if/when SRS is added.
Is there any possibility to type in accents? Switching from the keyboard to the mouse to get to the accent keys takes so much time 
Basically I would like to type “`+e” to get “è” etc. but that doesn’t seem to work.
(Am on Ubuntu 19.04)
Thanks for the feedback! Agreed switching from keyboard to mouse is not ideal 
Typically I like to just switch my computer keyboard to my target language - in this way I get the added benefit of learning the Spanish keyboard layout. If that’s not an option - are you able to type “`+e” to get “è” like you mentioned on other sites / in other applications? If the issue is just with TG2 I’ll try to debug, otherwise I’ll see if I can’t get some accent character shortcuts added.
Hi Mike!
Thanks for the fast answer.
My keyboard is set to “Generic International”, I’d rather not use the highly confusing french layout
I then tried typing " `e" in these applications/sites and it all gave me “è”
Thanks for the followup! I pushed up a potential fix - could you please let me know if you’re still seeing the issue?
Another short question:
The audio on TG^2 on my system (Firefox on Ubuntu) is really bad.
The audio on Clozemaster (on the same system) is way better.
Do you use different/multiple text-to-speech engines for the two websites?
Is it possible for users to set the text-to-speech engine?
Thanks a lot in advance!
Thanks for the question and feedback!
Do you use different/multiple text-to-speech engines for the two websites?
For the moment, yes - TG2 relies on your browser/system TTS. Clozemaster does this for most languages as well, but it also has higher quality TTS for some languages that comes from Clozemaster.
Is it possible for users to set the text-to-speech engine?
At the moment you can only select the voices, but would have to switch browsers to see if any other voices/languages are available. Chrome tends to work best.
I’d like to be able to have the TTS coming from TG2 at some point in the future so we can have the higher quality voices consistently and cross browser, but that can get expensive and I’m still working on figuring out the business side of things. For now the TTS is surprisingly good in some browsers/systems (and surprisingly bad in others 
).
I’d be curious to hear if Chrome on Ubuntu works better. Will keep you posted if/when we can make the switch. Thanks again!
Greetings! 
Given this sentence:
It is split into 2 sentences, because of the “.” after “9”. However, it is just one sentence, since “9. dakikada” means “in minute 9”.
Here’s a possible solution, which most likely can be improved upon.
When detecting a “.”, if all previous characters are numbers, it should not break.
This wouldn’t solve every occurence of “{{number}}.”, but woud atleast solve those occurrences that happen at the beginning of the sentence. This should be safe for most cases, but it could however interfere with someone who’s using the tool to practice numbers in isolation, which would be a downside to this approach.
On the other hand, a different approach would be to have a checkbox option when importing the article, in which we can specify that we don’t want a seperate sentence if the last character is a number. If enabled, this one could be slightly problematic in the sense of joining a couple of sentences together whenever they end with a number. However, since it’s a parameter that would be turned off by default, in general, most users would leave the setting as is, and thus, not interfere with usual imports.
It’s also worth noting that from my experience, I wouldn’t mind so much ocasionally translating 2 sentences in one go, rather than having one sentence split, and now dealing with 2 machine translations that make little sense, having to unify the sentences and delete one of them.
Hello!
Thanks for the feedback!
When detecting a “.”, if all previous characters are numbers, it should not break.
I’ll see if I can get something like that implemented. The checkbox option might be a bit trickier / more complex. I may also just do it for Turkish for now. Curious to hear if anyone else has run into this issue and with which languages if so.
Thanks again!
Tried importing the following article:
It’s still loading after 10 minutes. I suspect that there could be a bug, since the article is not very long.
I seem to be having problems with the review feature. I can type the translation but it doesn’t score points. Before it was intermittent but now it appears that it happens all the time both in favorite sentences and played sentences.
@holachicos2 thanks for letting me know! To be sure - it works as expected, but it’s just that the points/words aren’t added to your totals as expected? Or is there some other issue?
@Punk thanks for letting me know and apologies for the slow reply! Will work on reproducing and fixing.
