Thanks for making this awesome extension! I love it a lot.
I have a suggestion. In the language I’m studying (Arabic), accurate subtitles are often not available. Luckily, LLN (and LLY!) is still a great help in those cases, because it makes it super easy to replay the line until I’m pretty confident that I can make out all words. Right now, I’m transcribing along in a text editor for lines I’m having difficulty with. But maybe a transcribing option could be added to LLN itself?
A transcribing feature could also be used to help out other users, if you can share your transcriptions with the community. Perhaps in the app itself (you can select “show community subtitles” and people can review the accuracy with an up-voting system?)? Not sure how hard/easy that would be implement though.
Anyway, please keep up your great work and I hope you’ll find my suggestion useful!
Hi. You are probably referring to the shows that are not originally in Arabic. In these cases, subs usually don’t match dubs audio, and that’s why at the moment we recommend using shows in original language for LLN. We have a plan to expand this though in the near future.
No, this is not what was meant. Shows and movies in Arabic are typically performed in a dialect but the Arabic captions are almost always in Modern Standard Arabic. As a result, there are vast differences between what is actually said and what is in the subtitles. However, I noticed that for the Arabic Netflix show “The Writer,” Lebanese/Syrian captions are available through LR and they are pretty accurate. How does this work? And which shows have this option? When I checked the Jordanian show, Al Rawabi School for Girls, the Arabic subtitle options are Arabic [Original] ASR and Arabic Audio Description. What’s the difference between the Arabic subtitles that Netflix provides to all subscribers and the Arabic [Original] ASR subtitles available to LR Pro subscribers? Thanks in advance!
This post may help you better understand this feature:
Basically, this feature processes audio into subtitles through speech recognition technology. It’s not perfect, but it can be useful if people want their subs to “match” their audio more accurately:
As well as ASR stands for “Automatic Speech Recognition”.
Thank you! I saw this but am still confused because “Automatic Speech Recognition” implies that the subtitles would be based on the audio, right? But for Al Rawabi School for Girls, I noticed last night that the ASR Arabic subtitles were different from the Modern Standard Arabic Netflix-provided subtitles but also quite DIFFERENT from the audio. For example, at one point, the ASR subtitles used a very different and different-sounding word for “but” from what the actor actually said. Both words mean “but” but no one would hear the actor and think that they said the other word… And there were many instances like that. It’s like the meaning of the audio is being translated into colloquial Arabic without regard to the actual audio. So strange…
Good catch! It might be that model that the ASR is trained on. I’ve noticed something similar as well for Korean — the word they choose is actually different from the word that was actually used just because it makes sense to the model or because it sounds similar to the word used, etc. The ASR subs are not perfect. Looks like they’re showing what they (the model it’s trained on) believe fits — at least from my understanding.
This is shows the amount of data/the datasets their ASR model has access to in each language, I believe — someone please correct me if I’m wrong here (and add to this if you have more info! )
)