Thank you Nicholas for your answer.
What I understood then is that importing an already compressed file to the bank and choosing a different compression technique will cause signal degradation and undesirable.
But in the project, the user is importing sound files from a variety of formats, I was thinking of feeding them all to the bank and choose one compression technique (such as MP3) but it seems that this is not a good idea.
Do you have any idea how to manage this? Thank you so much.