A question about the new audio API

So, not entirely sure if this has been asked before, but what is the (most efficient) way of adding effects to an audio stream? Is that to pass the samples through a stream to convert it to the native device format and then to pass them through an effects engine (e.g. SteamAudio) and then to put those new samples into the audio stream, or is there another way? The new audio API is fantastic, just am curious about doing it this way since I could pull in Miniaudio and just use that (as an example) but it seems rather wasteful when SDL already has an audio system (though in the future it would be nice if it could see some additions like Doppler, just basic audio effects).