Unreal has a bunch of non-destructive features when it comes to lowering quality of assets for smaller packages. For example, I can force a mip of a texture and have it package at that size, but I can always come back and raise it later without having to back up and re-import any external assets. Very nice for handling games that are release on multiple platforms like PC and Mobile
I was wondering if this was possible to do something similar to the sample rate for audio? Could I lower the sample rate of my audio from 44100 Hz to 22050 Hz to package for mobile to get the size down, and then move it back up for the PC build?
I can do it manually by lowering them all externally by hand, backing up the high res versions, and then re-importing for every audio asset I have. It will take forever, so I just wanted to make sure this is my only option before going for it.
Also, I play tested the 8000 hz windows build (can’t deploy to android at the moment) and the sound is very obviously not at 8000hz; the audio is still nice and crisp
That is very weird. The sample rate change was working for me when I used it for Oculus Quest builds. I will try to dig around more and see what I can find. In the meantime, if you are able to solve this issue, do post it here!
I can confirm the pitch issue, both a higher and lower pitch. This also happens in 4.22. I came across this when trying to upgrade my project from 4.21 to 4.22 and also 4.23. Setting the sample rate to 48000 is what fixed it in my case. I don’t know why this affects pitch.
Hello, all. I want to chime in to add information. Changing the sampling rate in Unreal is not the same as performing a sampling rate conversion. For example, if you have an audio file that was recorded at 44.1 kHz and play it back in a system that uses 48 kHz, it will pull up the audio to play at that selected rate. That will affect pitch and duration. The reverse is also true. If you don’t want that to happen, either perform a sampling rate conversion on the file in question before importing it, or record the file at the destination sampling rate in the first place.
If you can prepare your audio assets at the destination sampling rate, that’s preferable over a sampling rate conversion, which although not catastrophic, affects sound quality by a small amount (although far less than Vorbis compression).
Also: when you change the pitch of an audio file in the usual way in Unreal, you are altering the sample playback rate. That’s how it works. It’s also the reason you get a formant shift accompanying it. If you need to manipulate pitch independently from a clip’s duration and/or formants, you’ll need to use software like Melodyne, Little Altar Boy, etc.