AudioKit: how to perform frequency modulation for AKPlayer - ios

Question

I'm trying to perform frequency modulation on a signal coming from AKPlayer, which in return plays a mp3 file.
I've tried to work with AKOperationEffect, but it doesn't work as expected:
let modulatedPlayer = AKOperationEffect(player) { player, _ in
let oscillator = AKOperation.fmOscillator(baseFrequency: modulationFrequency,
carrierMultiplier: player.toMono(),
modulatingMultiplier: 100,
modulationIndex: 0,
amplitude: 1)
return oscillator
}
Has anybody an idea how to get the mp3 modulated?
Unfortunately, the AudioKit API is not so well documented .... there are a tons of examples, but they all deal with synthetic sounds such as sine, square waves etc.

Answer 1

I took the time to create a working practical example to help you #Ulrich, you can drop and play if you have the playground environment available, or just use it as a reference trusting me it works to amend your code, it's self-explanatory but you can read more about why my version work after the code TLDR;
Before <audio>
After <audio>
The following was tested and ran without problems in the latest XCode and Swift at the time of writing (XCode 11.4, Swift 5.2 and AudioKit 4.9.5):
import AudioKitPlaygrounds
import AudioKit
let audiofile = try! AKAudioFile(readFileName: "demo.mp3")
let player = AKPlayer(audioFile: audiofile)
let generator = AKOperationEffect(player) { player, _ in
let oscillator = AKOperation.fmOscillator(baseFrequency: 400,
carrierMultiplier: player.toMono(),
modulatingMultiplier: 100,
modulationIndex: 0,
amplitude: 1)
return oscillator
}
AudioKit.output = generator
try! AudioKit.start()
player.play()
generator.start()
Find the playground ready to use in the Download page ( https://audiokit.io/downloads/ )
As you can see, apart from declaring a path to the mp3 file when initializing a new AKAudioFile and passing to an AKPlayer instance, there are three steps that you need to occur in a certain order:
1) Assign an `AKNode` to the AudioKit output
2) Start the AudioKit engine
3) Start the `player` to generate output
4) Start the generator to moderate your sound
The best way to understand why this is to forget about code for a bit and imagine patching things in the real world; and finally, try to imagine the audio flow.
Hope this helps you and future readers!