Apple Music uses a series of signals to learn your low-pitched tastes. The many apparent one is that we can categorically Love or dislike a sold lane by flagging it in iTunes or a Music app. You can also do this on HomePod, by observant ‘Hey Siri, like this track’ or ‘Hey Siri, dislike this track’ while it is playing.
But it also uses a series of algorithms to automate a learning. For example, if we play a sold lane often, it will assume we like it. And simply personification a lane and listening to it all a approach by (as against to skipping it) has some grade of influence.
All of that gets cryptic when more than one person controls it …
My partner and we have rather opposite low-pitched tastes. We do have common genres and artists we both like, though we have song Steph dislikes, and vice-versa. We also let guest control it when they visit, and their low-pitched tastes might again change from mine.
HomePod has a resolution for this: we can tell it to possibly use or not use your listening history. But that’s not a good solution.
If we leave listening story on, afterwards my Apple Music comment gets shabby by a tastes of my partner and a guests. If we switch it off, afterwards my comment no longer learns from my listening habits on HomePod, and given that’s many of my listening these days, that’s a large problem too.
In theory, there’s a third option, that is to switch it on and off repeatedly, depending on that of us is determining it during a time – though that can barter behind and onward mixed times a day, so isn’t unequivocally practical.
The ideal solution, of course, would be for HomePod to offer multi-user support. Associate dual or some-more opposite Apple IDs with it, and brand ourselves by voice. Then have HomePod commend a voice and refurbish a suitable listening history. In that way, it would also do things like review a scold content messages and check a scold calendar when anyone asks.
But if Apple can’t do that most during this stage, afterwards a good initial step would be simply to commend a voice of a categorical user. If that user plays something, refurbish their listening history; if anyone else does, don’t.
Distinguishing all voices might be formidable in some homes. You could, for example, have dual sisters with really similar-sounding voices. But for a simplest probable box of a male/female couple, it ought to be trivial. And not too formidable to heed same-sex couples, and adults from children.
Is this a underline you’d like to see? Please take a poll, and share your thoughts in a comments.