As I can see each line of subs (@timecode) is processed independently, without surrounding context.
This makes translation very rough.
The much better results could be achieved translating text as a whole.
(But I understand, that it can be difficult to set proper timecodes then)
As I can see each line of subs (@timecode) is processed independently, without surrounding context.
This makes translation very rough.
The much better results could be achieved translating text as a whole.
(But I understand, that it can be difficult to set proper timecodes then)