New on iOS17 we can control the amount of 'other audio' ducking through the AVAudioEngine. Is this also possible on AVAudioSession? In my app I don't voice input, but I do play voice audio while music from other apps plays in the background. Often the music either drowns to voice, if I use the .mixWithOthers category, or it's not loud enough if I use .duckOthers. It would be awesome to have the level of control that AVAudioEngine has.

Thank you for this new API. Today, when using AUVoiceIO, voice gets processed ahead of rendering to ensure the echo canceller is capable of filtering it out from the input. Will other audio be processed in the same way? For example, rendered as mono in a 16kHz sampling rate? I'm asking because I'm wondering if this API will unlock the ability to use wide-band, stereo, high quality other audio (for example game audio) simultaneously while using voice. Thanks!