Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Spatial Audio

lefsetz.com

Post reply on HN
Print view View original

31–40 of 138 posts

Re: Spatial Audio

#31

Post by doctorhandshake »

matwood wrote: post #17
allears wrote: post #3

Earlier quoted context omitted.

Lots of people like gimmicky things better, just like lots of people like to turn up the bass and treble all the way. But lots of us (and I'm a musician and audio engineer) appreciate high quality balanced distortion free audio, and want to hear the music the way the original artist intended. Just because you can make sounds bounce around in space doesn't make the music better, especially if the original recording wa…

> and want to hear the music the way the original artist intended Curious how you know what the artist intended. Did they intend for the track to sound like you're at a stadium show or sitting in a small music studio with headphones on?

There is a notion of ‘reference’ in audio and visual information. You can get reference speakers, headphones, and monitors - their role is to reproduce the information without ‘color’. They present ‘faithfully’. This is what is often meant by ‘listen as X intended,’ although to be honest the interest of many audiophiles is to set up a listening environment that does impart some color while remaining faithful to the artistic choices in the mix of the original music. And mixing _is_ an art.

The article suggests, and I agree, that by making what are in effect unauthorized remixes of these works, Apple are interfering with what the artist intended.

Eddie Cue, Apple SVP of Services (in charge of Apple Music) is quoted in the Verge article [1] on this as saying, “This requires somebody who’s a sound engineer, and the artist to sit back and listen, and really make the right calls and what the right things to do are. It’s a process that takes time, but it’s worth it.” Of course you note he says this requires the artist, not present in Marvin Gaye’s case and likely most others, and time, which it is implied is not being given to this process in the article above.

The visual analogy in the article is apt - imagine Apple were revisiting their movie library - including old classics like The Maltese Falcon - and applying a process that makes them viewable in simulated 3D — but that, even further — if that person tries to view it with equipment that doesn’t support the head-turning 3D effect - they see it in an _altered_ version of 2D from the original. And this now becomes canon for a certain group of people - this version that is a byproduct of an effort to reverse the effect of a series of creative decisions made to make a derivative work of the original, done without, in this analogy, the input of the director or cinematographer.

[1] https://www.theverge.com/2021/6/9/22525028/apple-music-spati...

Re: Spatial Audio

#32

Post by Cybotron5000 »

phiresky wrote: post #27
Cybotron5000 wrote: post #7

Spatial audio per se is not a scam, but its effective simulation is actually potentially rather complicated and I too worry about its successful implementation by Apple… Funnily enough, I had plans for a system to implement spatial audio using smartphones/headphones originally back in 2006 and more recently wrote an app for iOS trying to implement it (…prototype/demo. that I made worked pretty well - used what I lear…

From what I understand, to really get spatial audio from stereo headphones, you need to use a HRTF (head-related transfer function) specific to the person. There's an open dataset of 50 different HRTFs [1] and a long video to compare them [2]. For me personally, only 2 of those 50 samples actually vaguely sound like audio is ever coming from in front of me. Left, right, and behind mostly works, but for most samples w…

You’re absolutely right that the most effective simulation would use a custom HRTF for each person, as the shape of the head, shoulders, pinnae of the ear etc etc. all play some part... There are some headphones that attempt to go some way towards this by taking pictures of/otherwise ‘scanning’ the users ear (eg. https://electronics.sony.com/360-reality-audio https://mysofa.audio/ …paper on the subject: https://www.aes.org/e-lib/browse.cfm?elib=19855 …interestingly, for Apple’s future plans perhaps, ear canal used for biometric validation: https://dl.acm.org/doi/10.1145/3351239 ) The strongest ‘cue’ is probably the ‘interaural time delay’, which is easy enough to fake, but as you say, without fixed sources (imagine the calibration/setup of a VR headset) and subsequently taking into account that people actually turn/tilt their heads to better appreciate the direction a sound is coming from, as apparently Apple does relative to the device the tracks are playing from: https://appleinsider.com/articles/21/05/18/apples-spatial-au... (in order to take best advantage of what you experienced where we can better locate left/right than forward/back...), the efficacy is reduced. Generally we are also much better at localizing sounds in the lateral plane than the vertical one (…evolutionarily perhaps this was more important in order to avoid being eaten by sabre-toothed tigers or something? :) There are other cues deriving from ‘head-shadow’ (somewhat like an eq.), reflection from shoulders and even reflections from the acoustic environment and so on that are subtler/more complicated… Traditionally dummy heads with anatomically ‘average’ characteristics and binaural microphones in the ‘ears’ have been used to make binaural recordings, but if you were to venture out with binaural microphones in your ears and then later play back the recordings the effect can be startlingly effective… [as an aside, when I posted the above comment in this thread a couple of days ago when it was first on HN, the title of the post was ‘Spatial Audio is a Scam’ :) ]

Re: Spatial Audio

#33

Post by defaultname »

Related question, but does spatial audio actually send Atmos separate audio streams with positioning data to the client? How does this impact data sizes?

I have listened to some of the SA samples and they sound good, but I still don't get the technology. If it's sending separate streams and spatially mixing them at the client, it's still ending up with two channels of audio, so why would it differ from normal stereo audio which can be mixed in such a way? And if it doesn't, then it's just a production process?

Indeed, it reminds me a bit of Q-Sound of the Madonna Immaculate Collection. Still the most impressive stereo headphone experience I've enjoyed.

Dolby Atmos makes loads of sense for oddball speaker arrangements -- you tell it that you have three speakers layered above, five spread out across the back, etc, and it can process to that arrangement optimally (versus say 5.1 that was geared for a very specific arrangement). I don't understand how it is relevant for headphones, which is specifically what it is targeting in this case.

Re: Spatial Audio

#34

Post by Cybotron5000 »

tinus_hn wrote: post #12
Cybotron5000 wrote: post #7

Spatial audio per se is not a scam, but its effective simulation is actually potentially rather complicated and I too worry about its successful implementation by Apple… Funnily enough, I had plans for a system to implement spatial audio using smartphones/headphones originally back in 2006 and more recently wrote an app for iOS trying to implement it (…prototype/demo. that I made worked pretty well - used what I lear…

As far as I remember it’s a patent minefield

If you can remember any further info. about this I’d be very interested.

Re: Spatial Audio

#35

Post by falcolas »

quenix wrote: post #22
tpush wrote: post #11

Earlier quoted context omitted.

You’re downvoted, but enjoyment of music is obviously subjective. There is no “wrong” way to listen to music (or anything really).

Exactly. This thread is mindboggling. Taste in music is purely subjective. If users enjoy listening to Spatial Audio, who are you to criticize it as a scam?

Because everyone using Apple music is forced to listen to it. Whether they like the remixes or not.

Re: Spatial Audio

#36

Post by bristleworm »

According to Apple, wireless headphones like the AirPods do support spacial audio. No idea where the author got the information they wouldn't.

Apart from that, I get how an audio engineer would think that way, but for me, and I suppose most music listeners, what's important is that the sound is great. And personally I really like spacial audio.

Re: Spatial Audio

#37

Post by kall »

onethought wrote: post #23
allears wrote: post #3

Earlier quoted context omitted.

Lots of people like gimmicky things better, just like lots of people like to turn up the bass and treble all the way. But lots of us (and I'm a musician and audio engineer) appreciate high quality balanced distortion free audio, and want to hear the music the way the original artist intended. Just because you can make sounds bounce around in space doesn't make the music better, especially if the original recording wa…

Do you end up in a divide by zero situation when the artist intended you to turn the bass and treble all the way up? :)

I think I remember some tracks where the artist is explicitly instructing me to turn up the bass (and make the speakers pop).

Re: Spatial Audio

#38

Post by Cybotron5000 »

defaultname wrote: post #33

Related question, but does spatial audio actually send Atmos separate audio streams with positioning data to the client? How does this impact data sizes? I have listened to some of the SA samples and they sound good, but I still don't get the technology. If it's sending separate streams and spatially mixing them at the client, it's still ending up with two channels of audio, so why would it differ from normal stereo…

This article does an okay job of explaining the idea: https://appleinsider.com/articles/21/05/18/apples-spatial-au... Apparently the sound source positions will be calculated relative to the position of the iphone/ipad/mac… ..Yes, I’m guessing it will potentially greatly increase the size of audio data downloaded/streamed to the listener, as the spec. for Atmos can have up to 128 different tracks + metadata: https://en.m.wikipedia.org/wiki/Dolby_Atmos

Re: Spatial Audio

#39

Post by astrange »

donbrae wrote: post #29
matwood wrote: post #17

Earlier quoted context omitted.

> and want to hear the music the way the original artist intended Curious how you know what the artist intended. Did they intend for the track to sound like you're at a stadium show or sitting in a small music studio with headphones on?

I imagine the artist approved a (stereo) mix they heard on decent studio monitors and maybe stereo headphones too. I’m not sure how they’d feel about someone else then creating an atmos/spacial audio version without their involvement. Mixing is integral to the artistic part of making recorded music. I’d be interested to know whether the tracks appearing on Apple Music are being created from the original tracks/stems…

They are remixed by the original engineers (or at least, an engineer.) Some of them are mixed better than others.

Re: Spatial Audio

#40

Post by basisword »

This sounds like the kind of thing someone would write when the world switched from mono to stereo. Quite a few parts of it are completely wrong too not to mention insulting ("... more music is being made by individuals in bedrooms, home studios, on a budget. They have neither the equipment nor the skill to mix in Dolby Atmos").

Apple has done quite a poor job showcasing it so far. Some of the tracks they've put in their playlists sound like crap. On the otherhand some of them sound fantastic. Jazz in Atmos is wonderful. It's also important to note that 'head tracking' comes with iOS 15 and in my experience improves spatial further.

A lot of music made for stereo won't remix well. If you listen to most of the rock tracks they don't sound good because they don't have a lot going on. 3 or 4 instruments + 2 or 3 vocals tracks aren't going to take advantage of spatial. Listening to some modern pop with lots of synths and layers upon layers of backing vocals, the experience is much better.

Atmos/spatial is a new tool like any other. People will write and record songs that work for it and take advantage of it. They'll use it creatively. That's what will be interesting. Most remixed songs from the 60's aren't the least bit interesting but the possibilities for the future are.

I could be wrong and it could end up largely ignored, time will tell. But there is the opportunity for a lot of creativity regardless of what cranky old music critics think.

Post reply on HN