Target language IDs include the source language
I assume I did this to simplify the flow and it seems all translation services we tried until now supported translating into the same language (e.g. from English to English) but Google translate does not which is why I was made more aware of this. I wonder if this is a common programming practice (see libretranslate) or if it might result in some weird mishaps (no idea what this could mean).
Two options:
- Keep this as is (maybe avoid sending a request for no reason to the respective service - as done in the Google translate adapter)
- Save the transcription somewhere beforehand and remove the source language from the target languages (need to figure out where and how but more importantly if this would not cause issues for cba podcasts or other clients)