Is it a bad idea to make a dictionary of pretty much sound cues?

Hi everyone, through the development of some project for vr, I need to make a system similar to asset localization to store dialogues of 4 characters in 4 languages, so I was making a dictionary that has a text as the key, and a sound base as the value, but then a thought came to my mind, I´m going to make this dictionary with around 400 sound cues inside of it, and I´m going to call it in a function, so the question is, when I call this function, Unreal is going to load 400 audios to memory? Or as I was thinking, It is going to load 400 references to those audios through the sound cues? Because it is such a big difference and also a big problem if it loads 400 audio files to memory in a mobile device. Any help would be appreciated, thanks.

If you are planning on using many assets then try using a soft object reference for the sound cues. You just have to load up the needed cue before use (Async load asset or load asset blocking).


You probably want to make a DataTable of references to them, which should force t hem to be packaged, but make all your references in code to them be soft references, so they aren’t force loaded until you need them.


That helps a lot!, I’m going this way. Thanks!

I still have that doubt, is it more memory efficient to use sound cues instead of sound waves? Or it will load the asset in memory anyway? Just for further knowledge.