thatDudo
thatDudo
There is not much, you can obviously choose the bigger translator models if you choose to use offline translators but otherwise im not sure what might cause the bottleneck.. Maybe...
Perhaps what is currently most impactful is single thread performance, which explains the 20% of cpu usage. So making the batch translation multi threaded might be well advised.
> @thatDudo coud you move this issue to disscussions I think that feature can only be enabled by the repository owner, if he wishes to
We have a special textline merger that merges lines of detected text based on their proximity and size difference. It may have problems with working the same on different image...
 Its still not perfect but at this point I think we will need to add some kind of text bubble detection like in --manga2eng to fix it.
Do you mean that papago isnt handling two different input languages correctly? We had the same issue in the google translator so i can just do the same thing we...
I should probably keep this open for a few days in case someone knows what in particular went wrong. But i think you mightve not set your api keys correctly...
While I doubt the necessity of outsourcing the logic to an external library I think it might be more intuitive to make the batch mode the default, though I dont...
It did it automatically because of the pr name
Would appreciate it if you could tell me if there are still problems with the rendering. Maybe I should disable hyphens ("-").