I've tried "once upon a time there was a king" (the first common words in a lot of fairy tales), which in Romanian should have been translated in five words - "a fost odată un împărat". This tool gives me "cu mult timp în urmă a Trăit-o a fost regele" - double of that amount (and I stop before getting into comments about the translated content itself)!
Thanks userulluipeste! It's great to get feedback, seems Yandex API isn't the best, very interested if there's any others people have had a good experience with?
Thanks for sharing!
(or... Mübadiləsi üçün təşəkkür edirik!)
My one thought is that string length in pixels is more important than character count, but you sidestep that issue by showing all the results. You can easily see if one is longer even though it has fewer characters.
The Chinese for "hello" was sort of correct. It returned 喂 on the bottom of the list, but that's really "hello?" like answering the phone, not 你好 which is actually hello. I was surprised by its selection there.
You might want to look into actual horizontal space. Visually the Japanese and Chinese translations takes a lot more space than the character count would have you believe.
Can you add an option to select the input language? For example if I submit a Japanese phrase, all the languages give identical results, which is the phrase itself.
Yep, on the feature list, currently it's limited to en -> Other only. Also looking at selecting the target languages so you only see the languages you want.
I just played with it for a while and the German translations often were pretty bad ("Essen sie nicht, dass" for "Don't eat that!", when "Essen Sie das nicht!" or "Iss das nicht!" would be correct)
Underneath it is currently using the Yandex API, mainly due to it's generous free tier. Can't speak to the quality myself, but it does the job as a quick tool for simple translations.