Developing critical listening skills


I’m not really an audiophile, but a long time music fan who values quality gear as a way to enjoy the music I love, so please forgive me if this seems naive. I’ve been experimenting with a lot of new gear lately, and with different resolution files, trying to see what differences I can detect. When evaluating hi-fi equipment, I have a good idea of the things to listen for, but find it very difficult quantify and compare differences. For instance, I just added new cables to my desktop system, and I think they sound better (wider soundstage, and more natural, less forced presentation), but I have a hard time identifying differences in a quantifiable way and really don’t know whether it is just an optimism bias. I can’t accurately remember how the sound was specifically different. I’ve always just listened to music on decent systems, but never tried to develop my critical evaluation skills: actually developing a systematic way to isolate, identify, contrast sonic differences. All the guidance I can find is very vague and general. Things like "spend a lit of time listening closely", or invest in the right hardware. I’ve already done both in spades. Are there some specific sort of reliable, audible tests that can be performed to build my skills? Any guides? I just purchased the Chesky Ultimate Demonstration Disc, and Sheffield Drive and A2TB Test Disc.

Similarly, trying to AB test files, and see if I can really hear a difference between 44.1/16 and a 256kbps file derived from the original, I honestly have a hard time. What should I be listening for? After a lot of listening to the same track, I think I’m starting to hear differences in the bass guitar, where the image a little smaller, and less resonant in the compressed file. Also, the cymbals are a little more sibilant, and with less depth and decay. But it is very subtile, and not too successful in an A/B test. Specifically what parameters should I be listening for (and how to I isolate & memorialize these characteristics repeatably) to start to build my listening?
svenerik
Post removed 
Reminds me of back when I was trying to figure this all out and not being able to hear the difference between two CD players, DACs, etc. Spent quite some time being really frustrated, because the whole time there was this feeling of something being there, I mean really being there not just imagining it, only I couldn't quite put my finger on it. 

Only real help I got, one time one guy at Definitive Audio said listen to the way things trail off. Like cymbals, instead of ting they go tiiinnnnggggg. Sort of helped. Not much.

Just listening to a lot of stuff, that alone will not do it. And you can forget about "quantifiable" or anything like that. Total waste of time. What has to happen is you struggle a lonnnng time doing just what you're doing now, trying to put words to what you're hearing.

The hardest part to explain is these things go hand in hand. There's even a whole school of thought that says you cannot even have an experience until you have the word for it. Sounds impossible. Yet it sure seems to be true. For sure, until I developed the vocabulary I really had no real idea what was what. 

In my case, I can still remember, there was one day when I had this epiphany. I had been listening to music at one store, came home and was playing a track from the XLO Test CD (Michael Ruff, Poor Boy) when it hit me, THIS is the quality I've been trying to find the words for!

Now today I know enough to say for certain it was the more natural balance between attack, harmonic development, and decay of each instrument. The lack of grain and glare. Dynamics that were full and natural not etched or hyped. Real true extension instead of a tilted up top end pretending to be extended highs. 

I could go on and on. Today I can go on and on. Back then though, just the glimmering of a few words.

That's the difference. You will find people who will downplay this. People who will try and tell you, you will know it when you hear it. To them I would ask, Know what? If you can't tell me what it is, how do you yourself even know? Odds strongly favor: you don't. 

There's a reason we read reviews. Reviewers, its not that they hear any better than anyone else. Its that they KNOW what they hear, and we know they know because they're able to tell us so. With words. Crude tools, only ones we have.

Best reference I know is Robert Harley's Complete Guide to High End Audio. Harley has a whole section devoted to audio terminology in which he describes just what to listen for, and why, and what it means. This all goes hand in hand with understanding all the various components, right down to the components within the components- the caps in a speaker, for example.

Because of this, because of the fact that everything matters, its not even really necessary to go and listen to a lot of different systems or components. Of course you need to do this when it comes to building your system. But in terms of learning to listen you can do this just fine with what you have right now at home.

Every recording, for example, is a window into a whole entire recording chain. No two sound the same, because no two are recorded quite the same. Read your liner notes. Pay attention to any recording details, especially the mastering engineer. If you have some different recordings with the same engineer try playing them one after another. Then compare them with recordings by another one. If you have any Doug Sax you hit the motherlode in terms of liquid smooth tubey magic. If you have any on the AudioQuest label, those all use AudioQuest wire throughout the recording chain. You can hear it in the recording- once you learn what to listen for.

Barely scratching the surface here, but more than enough to get you going. Even just asking shows you are on the right track, more so than most. Keep at it. One day you may be like me and find much to your amazement what once seemed near impossible has become child's play.
Who cares? Unless you are going to get a job making audio equipment or writing for a magazine, it’s a waste of time. For the non-professional, critical listening is a disease. I listen to enjoy the music, not to nitpick my equipment until my bank account is empty. If you like the way something sounds, that’s good enough. As a dabbler in live sound and mixing, I am more prone to criticize the recording and mixing engineer’s choices. There is a much bigger difference in the quality of recordings than there is between competently engineered components.

p.s. A pretty big chunk of what audio reviewers write about is imaginary.  Don’t feel bad if you don’t hear it.
Wow @rwortman that’s rough. For many people, like myself, it’s not imaginary. There is a big difference and it helps me connect with my favorite music even better. Orchestral and jazz in particular. Some music is hard to reproduce. For instance, growing up I hated older orchestral recordings. It wasn’t until I started having a more revealing system that I really appreciated those recordings and now love them. You may not like that, but no reason to put other people down, like reviews, who I find to be very honest, and offer insights that are invaluable as I continue to explore new types of music. 
Post removed 
https://harmanhowtolisten.blogspot.com

steakster, when Harman (Floyd Toole) did their big speaker/listener experiment a few years back, the reviewers who did participate performed rather poorly. That is obviously not indicative of all reviewers. I am glad you found one of the diamonds and he was willing to share the experience with you. 
@svenerik
A few simple pointers might help
(...)trying to AB test files, and see if I can really hear a difference between 44.1/16 and a 256kbps file derived from the original, I honestly have a hard time
It helps if you listen to loud music (because our hearing is "flat" at higher amplitudes), listen to s/thing you're familiar with (see elizabeth above); you also need a system capable of reproducing low-level detail, which I imagine your system is.
Listen for:
- sense of space between instruments and volume (not loudness, i.e. does the tenor sound like a tenor or a mouse?)
- 3-dimensional soundstage, wide and deep?- Bass: is it clear? Not overpowering, but clear- Do you hear low-level detail
- Do you get listener fatigue very quickly?
There are many other considerations, but these are the simplest I think.
Three ideas:

When was the last time you had your ears professionally cleaned?

Listening to music through headphones might help you
develop critical listening skills.  By "taking the room out of it,"
you can focus on "what you hear" more easily.  

Lastly, if you have another person to share your perceptions with,
his may be helpful.  "Do you hear this?" and similar questions and comments
can direct your attention towards making important discriminations...

Lot of good advice on this post. I second Robert Harley's book. After reading his chapter on becoming a critical listener a whole new layer opened up for me.

rwortman makes a good point too. The quality of the recording make a substantial difference. I listen to one song after another and it sounds like the music is coming from the center, then another song comes on and it's distinctly left channel/right channel.

elizebeth nailed it though; use a few songs that you know backwards and forwards for reference.
All you need to do is listen "it is like riding a bicycle" it is not rocket science there is no special skill required you were born to listen.
Floyd Toole's research at Harman would suggest otherwise. Here is how a variety of listeners performed (100% being best). I have to ballpark the numbers as it comes from a graph:

  • Selected and Trained listeners:  >90%
  • Retail Sales (audio): 35-40%
  • Audio Reviewers: ~ 20%  (but admitted small sample size)
  • Brand Sales and Marketing ~10%
  • Student (i.e. completely untrained) ~ 5%

The How to Listen blog/app I linked and the Harley book are good resources. I will warn you, being a critical listener is a double edged sword. Once you know something is "wrong" you can't ignore it any more. It can interfere with enjoyment.


roberttcan
Floyd Toole's research at Harman would suggest otherwise ...
Quite so! Those who seek statistically valid results from controlled, double-blind listening tests often provide some training for the test subjects in advance of the test. In my experience, this training typically includes demonstrating for the subjects exactly what difference or anomaly they are trying to detect in the test. Test organizers know there is a "difference," so the only question is: Is the difference reliably detectable? Without some guidance, even experienced listeners will often struggle.
@svenerik  Are you enjoying your music? Are you able to relax and let yourself 'be' in the moment during your listening sessions?


It's important to have two benchmarks, music that you know well played through a system you know well, and real live music.  When you audition a new piece, in theory you can then judge whether it then sounds better, worse, different or the same as the components you know, and the criterion for better should be it sounds more like what you hear as live music.
@svenerik Are you enjoying your music? Are you able to relax and let yourself 'be' in the moment during your listening sessions?
@david_ten Yes, listening and enjoying is something I've never had a problem with. I can say if something sounds good to me, or is somehow off or un-involving.  The skill I lack is being able to objectively evaluate & compare listening situations, and determine why something is off. Hence my desire to develop those skills.
@twoleftears i often hear of the idea of using live music as a benchmark. I can see how this would be of value when evaluating a live recording. The problem I have is that the studio recordings I listen to are typically not mixed to sound like like a live stage. The drum kit may be spread out in stereo, and while individual instruments can be pinpointed don't have the actual ambiance of a venue. So the recording never really sounded live to start with. And virtually all the live performances I see use sound reinforcement, so you are really hearing the house PA more than the instruments. Albeit the house PA at my favorite venue (The Baked Potato) is top notch.
The Harmon How to Listen is a useful site for learning how to . Also Audiocheck.net is great for testing a myriad of things on your equipment and your hearing capabilities and limits.  
@svenerik   First, kudos to you for being open to learning. Second, your questions are good ones and the responses posted are informative and will (hopefully) help guide you forward.

Adding to what has been shared, I suggest staying "tuned" / "alert" to your emotional connection and engagement to the music as a significant metric. We tend to deconstruct (emphasize deconstruction) in this hobby...therefore, the reminder to consider the whole.

I also recommend broad strokes first. One doesn't need intimate familiarity with recordings to assess elements such as pitch, tone, timbre, timing, etc. Use your own definitions (understanding) of these as a starting point.

Good luck. And make it fun.

When I mentioned live music, I wasn't thinking so much about soundstaging as the inherent timbre of instruments and the human voice, particularly unamplified ones.  If the studio recording distorts these (unintentionally) then it's not doing a good job.  In that sense several recordings* of unamplified instruments measured against the sound of unamplified instruments heard live is, to my mind, the best way of evaluating the reproduction of a given system.

* to eliminate the variable of how they got recorded in the studio

+1 @david_ten
+1 @twoleftears

- Visit a piano store. Listen to a customer playing.  Stand at different distances away.  Especially, pay attention to the high notes.
- Visit a Guitar Center. Listen to a customer playing a drum kit. Especially, pay attention to the cymbals, snare and kickdrum.