Self-testing and toys to help you learn about your music listening

March 23, 2023 9:58 AM

The Music Lab has tests for you to learn how good you are at discerning melodic discrimination and recall, mistuning perception, and beat alignment and more (previously). "This Is What It Sounds Like" offers compare-and-contrast samples to help you reflect on your taste in melody, novelty, realism, timbre, and other elements. Its links lead to further online tests and demonstrations.
posted by brainwane (37 comments total)

This is fun and challenging, I think I was doing pretty well (13/15) on the first two sections but I bumped the button on my phone and won't be able to play until later. It's really tough when both examples are very close to each other and the matching key or beat. The vocals especially remind me of the question a while back about "Is it just me or is Bette Midler always out of tune."

Tips: don't accidentally reload the page you will lose your progress! Also yes do use headphones and make sure you do have about 20 minutes.
posted by SaltySalticid at 11:01 AM on March 23, 2023 [2 favorites]

this was fun! I am not actively playing a lot of music these days (unless you count singing to my toddler), but I took lessons of various types from about first grade through the end of high school. I was a little depressed about my performance on the mistuning perception part.

I didn't use headphones, though, maybe that would've helped.
posted by dismas at 11:02 AM on March 23, 2023

"Is it just me or is Bette Midler always out of tune."

No, it's the band who are wrong.
(insert Principal_Skinner_rubbing_chin.gif)
posted by Greg_Ace at 11:12 AM on March 23, 2023 [1 favorite]

Ha, here's the question if anyone is interested. The good answers about how timbre can interact with our notion of pitch I think are well highlighted in some of the examples I got.
posted by SaltySalticid at 11:19 AM on March 23, 2023 [1 favorite]

Brainwane, you are a treasure!
posted by BlueHorse at 11:28 AM on March 23, 2023 [2 favorites]

Huh, I clearly don't have the advanced listening skills I've always thought I had.
posted by slogger at 11:30 AM on March 23, 2023

That was fun and a little exciting waiting for the final number (118, baby).

The beat alignment seemed the easiest. As a lifelong musician, I was surprised I did worst in the melodies. That felt almost like memorizing a string of numbers. (Any major/minor switch made it easy.)

The mistuning was tricky. Compared to the beats and the melodies, it was more judging on a continuum than clearly right or wrong. Sometimes the voice was buried in the mix, or the first one sounded ok until you heard the second was clearly better. I'd love to hear how people with perfect pitch do vs the rest of us.
posted by gottabefunky at 11:32 AM on March 23, 2023

It would also be great to be able to go back and re-listen to the ones you got wrong.
posted by gottabefunky at 11:37 AM on March 23, 2023 [3 favorites]

The melody quiz seemed like a "one weird trick" problem, where one melody would zig where the other two would zag (those are official music theory terms). Once I figured that out I was able to answer them more accurately.
posted by slogger at 11:43 AM on March 23, 2023 [1 favorite]

Yay, The Music Lab! See also this tone-deafness research from a few years ago. I think I also participated in an earworm survey with them at one point a while back, though that could have been some other online music research nerds.

I did pretty good on this, which I expected since I've done so much music over the years, but it was interesting to see where I did less well—I'm way out on the long tail on melodic discrimination but only a standard deviation or so out on mistuning and beat alignment.

I'm not shocked to not be better on mistuning, really; a lot of people tend to be bothered by subtle tuning and intonation issues that don't jump out at me, and on some of the specimens here where I was confidently correct I found that, well, yeah, but I didn't really care. I think I like too many musicians who are a little sloppy and habitually flat or real slow and casual about their portamentos to be as fussed as I technically ought to be able to be about it.
posted by cortex at 11:57 AM on March 23, 2023

As a lifelong musician, I was surprised I did worst in the melodies. That felt almost like memorizing a string of numbers. (Any major/minor switch made it easy.)

That one definitely had a lot to do with feel at times; I can pick up a melody pretty quickly but not, like, eidetically on the first go in general, and I found myself struggling with the desire to try and quickly map the literal melody the first time vs. relaxing and trusting that recognizing the presence or absence of accidentals or significant degrees in the melody would get me there. Which makes it less about recognizing the literal melody than about recognizing conspicuous key/modality features in the melody: ooh, only one had a minor third in it; ooh, two of them were strictly major pentatonic, the other had an extra note in there; etc. I think I'd do a lot worse looking for the odd man out if it were a pair of slightly different but basically tonally random "melodies" where I couldn't get a foothold, or if the specimens where from a folk tradition with microtonal degrees that I'm not as used to thinking about.

I also suspect I was more consistently confident in cases where two of the three specimens in a row had the same melody; give me ABA and it's just that much more mental acrobatics to manage and compare within short term memory.
posted by cortex at 12:06 PM on March 23, 2023 [2 favorites]

This was very fun. Starting with beat alignment was mean because I got a false sense of confidence.

About two thirds of the mistuning test were meant as a test for a new torture method I think. Yikes I'm glad they cut those short. The other third were really a crapshoot because either I couldn't tell what they were going for or there wasn't really a reference point behind the voice to tune it to. There was only one where I was SURE of the answer and I was wrong.

I was very bad at melodic discrimination. I feel like if they played the 3 melodies twice I would have done far better. I was trying so hard to tell if the first two were different, which I think I did, that when the third one came up I had both of the first two in my head and just mostly guessed.
posted by Clinging to the Wreckage at 12:07 PM on March 23, 2023 [1 favorite]

In which I learn that 66% of people perceive pitch better than me.

Does being fluent in a tonal language mean you ace the pitch perception test?
posted by aniola at 12:10 PM on March 23, 2023

I was slightly above average overall; did best at the tuning, then beat, then melody. This doesn't surprise me too much given that French horn is my primary instrument- you need a relatively strong sense of relative pitch (you play higher up in the harmonic series than other brass instruments, so the notes are closer together) and tuning (hand in bell messes with the tuning of the horn, which is good because you can make microadjustments but also bad because, well, you can make microadjustments), and then years and years of doing off beats in marches, with not a ton of playing the melody in things!
posted by damayanti at 12:37 PM on March 23, 2023

I was about a standard deviation above average, but only about average on tuning. Sometimes, it felt like the degree to which things were off might have been intentional until you hear the second one, but most of the ones I got wrong I was pretty coin-flippy about.
posted by dismas at 12:42 PM on March 23, 2023 [1 favorite]

The beat alignment seemed the easiest. As a lifelong musician, I was surprised I did worst in the melodies. That felt almost like memorizing a string of numbers. (Any major/minor switch made it easy.)

Some contrasting anecdata, I am not remotely a musician (I had three or four piano and guitar lessons my parents didn't care to keep paying for as a kid and that's it) and I did much better on melodic discrimination (122) than on beat alignment (107) or mistuning (102). I'm also very good at memorizing strings of words, because I visualize them rather than repeating the words themselves in my head. Similarly, I visualized the notes as patterns (I can't read musical notation, so it was more like a line graph) and compared those, which was easy enough. However I'm quite bad at memorizing strings of numbers because numbers exist on a visual line to me and it's really hard to contort that line to match e.g. 261473.

I'm actually pretty shocked by my results because I always thought I was pretty bad at the whole music thing. But I spend a lot of time around really musically inclined people, so I think my measure of average is skewed.
posted by brook horse at 12:48 PM on March 23, 2023

Also, most of the songs on that mistuning test were my jam. I wrote down some of the lyrics and am disappointed to find none of them seem to exist, which makes sense scientifically but :(
posted by brook horse at 12:55 PM on March 23, 2023 [2 favorites]

I got a 122 overall ("You did as well as or better than 92.88% of people"), with the beat alignment surprisingly the lowest (although I got one wrong by selecting the wrong answer because I forgot how the quiz worked). The mistuning one was stressful, though. Either it was obvious and painful ("gah, that one, that one, make it stop"), or there were a couple where I was like "they're both not great, and I can't tell if the sharp one was more sharp than the flat one was flat." For that matter I didn't like the melodic discrimination one because some of them seemed more like random sequences of notes than melodies, but I did score the best on it.
posted by fedward at 12:56 PM on March 23, 2023 [2 favorites]

(Full disclosure which is obvious if you followed the link to the Bette Midler thread: former professional musician with over 20 years of training).
posted by fedward at 12:57 PM on March 23, 2023

Tuning:124, Beats:108, Melodies:134. Fuggin' computer beep, humph.

Try the super listener test. You hear two quickly arpeggiated chords, but it's the whole scale in random order (not just thirds): a jumble. Then you identify which sound is which. That's a threadbare explanation, you have to hear them.

I'd like to know what they were. I thought I heard major versus minor, and then somethingsomething versus somethingsomething-with-a-flat-five...and I got them all correct. But I couldn't figure out what they were exactly.
posted by Rich Smorgasbord at 12:59 PM on March 23, 2023 [3 favorites]

As I expected, I did best on beat alignment (115) - all those years of Guitar Hero paying off, baby!

Mistuning (113) and melodic discrimination (104) were harder - I especially struggled on some of the longer melodic pieces.
posted by hanov3r at 1:00 PM on March 23, 2023

137 beat
120 melodic
116 mistune
I'm a bass player so it makes sense that I'm a little tone deaf!
posted by pepcorn at 1:23 PM on March 23, 2023

real lake wobegon problem with this test apparently
posted by dismas at 1:43 PM on March 23, 2023 [1 favorite]

101. Almost dead center in that bell curve. The first and last tests I did better on, and some of the mistuning questions were easy, but easily half were just a tossup for me. I think, though, that the more you take this test the better your score will be. Once you get used to the testing format you can absolutely raise your grade.
posted by zardoz at 2:17 PM on March 23, 2023

real lake wobegon problem with this test apparently

*posts picture of Garrison Keillor with a bunch of uneven clumps of red circles on him*
posted by cortex at 2:27 PM on March 23, 2023

real lake wobegon problem with this test apparently

For what it's worth, I didn't even finish the first test, because I turns out that I'm even more terminally incapable of following a beat than I always thought. I was always just guessing.

I've been playing an instrument for years, but I know why I'm only doing it in the privacy of my own home.
posted by sohalt at 2:49 PM on March 23, 2023 [1 favorite]

It's kinda bogus to use a sample with so much extrinsic noise for the melodic exercises. Especially on the short phrases. The action was louder than the notes! It also irritated me that on at least two of the mistuning questions, the "wrong" answer was enough out of tune I preferred it musically over the one that was actually closer. Now, as a solo act mostly, my playing can most charitably be described as rubato, so I'll own that beat alignment. Still, quite annoyed overall.

You did as well as or better than 65.54% of people.
Your Musical IQ is 106
Melodic Discrimination: 96
Mistuning Perception: 113
Beat Alignment: 108
posted by ob1quixote at 4:50 PM on March 23, 2023

I had my eyes closed during most of my first part,which was beat alignment, so didn't realize they scored you one if you were right or wrong until the second one. I started to do better once I got feedback! Also I wish there was more of a pause between each song, I felt hurried, like they were too close together.
posted by Carillon at 10:35 PM on March 23, 2023

For what it's worth, I didn't even finish the first test, because I turns out that I'm even more terminally incapable of following a beat than I always thought. I was always just guessing.

Same. I had the melody test first and I got tired of not being able to remember the first melody by the time I got to the third. I wish they were split up becuase I am interested in how I would do with beat alignment, but not enough to choke on my ignorance for 11 meldoy questions.
posted by dame at 6:54 AM on March 24, 2023

I did bad at the beat alignment part but I had a hard time hearing the underlying music through the beep track. 124 on melodic discrimination but literal "playing by ear" runs in the family so that makes sense. On some of the mistuning questions I felt viscerally ill when hearing the out of tune track. Maybe I have perfect pitch? 113, so maybe so.
posted by fiercekitten at 8:17 AM on March 24, 2023

real lake wobegon problem with this test apparently

I got a 95 overall. I think I got a little below random guessing for the out-of-tune recognition.
posted by paper chromatographologist at 8:53 AM on March 24, 2023

Got 123 overall. Worst subscore was beat alignment, which disturbs me because drums are my weapon of choice :-(
posted by flabdablet at 11:19 AM on March 24, 2023

That was fun! I’m like hopeless with music. Pretty sure I’m tone deaf, and I can’t keep a beat. But I got 115. 13/15 on the beep test too! Nice little confidence boost!

I can’t wait to see what my wife (music teacher) and son (orchestra/theater dork) get.
posted by Don.Kinsayder at 5:33 PM on March 24, 2023

Did anyone try the Analytic vs. Synthetic Listening test?
posted by brainwane at 10:37 AM on March 27, 2023

Sounds like a touch-tone phone. Apparently I'm an analytic listener.
posted by fedward at 12:20 PM on March 27, 2023

Oh, that's a great little aural Necker Cube. I can go back and forth depending on what I want to focus on; it's impossible to see past my own subjective interpretation but that honestly gets into a nice kind of meta-synthetic space for me where I'm interpreting it as a full on I/iii chordal movement with both the lowered bottom tone AND the added implied fundamental.

That tendency of the brain to fill in implied fundamental is a really fascinating part of music perception: instruments and compositions can in a subjective sense go lower in perceived tone than the frequencies actually being produced, by providing a set of overtones/harmonics that "belong" in our acquired sense of musical tones to a base note that is not actually present, to the point where the mind fills in the gap.

e.g. if you take a low C note on say an upright bass and look at its harmonic content, you'll see a typical mix of fundamental and harmonic tones: the low C note itself, the "fundamental", which is the whole length of the string vibrating; and then the first harmonic which is a C an octave higher: the two halves of the string vibrating around a still midpoint; and then a G above that, the second harmonic, with the string vibrating in three equal sections; C above that, two octaves about the fundamental; E above that by a major third; G above that; and then so on, with various higher harmonic bits getting more narrowly spaced and being a smaller portion of the overall sound.

All those various harmonic tones that aren't the fundamental note itself create a complex harmonic texture that is the timbre of the instrument. This is one of the core reasons why different instruments sound different, why a piano and a bass and flute and and oboe all can play the same note but sound immediately starkly different: the proportional mix of harmonics that a note on each produces is very different one to the next.

But part of the magic here is that, because we get used to hearing notes as not just a fundamental frequency but as a big mix of overtones, you can take a familiar note and timbre and subtract some of those harmonics and it'll still sound more or less like itself, just maybe a little weird and filtered. That includes removing the fundamental frequency: if you can still hear the first, second, third etc. harmonic of the note, the fact that that low low bottom tone isn't there doesn't stop you from perceiving it as that low note, because no other note has the same mix of harmonics on top.

One nice implication of this is that even on speakers with bad bass response, you are likely to still "hear" the bass notes to an extent, even if they don't have a satisfying thump of the literal low-frequency sound waves of that fundamental tone hitting you. Most of the musical information is still there, just implied; you can still decode it even with some information loss.
posted by cortex at 9:27 AM on March 28, 2023 [1 favorite]

I remember reading an article in an electronics magazine back in the 70s that gave an account of a speaker manufacturer (Philips?) that was experimenting with that effect to get amazing sounding bass out of speakers that were, for their time, tiny.

If I recall correctly, the technique involved applying a certain degree of deliberate harmonic distortion to everything below about 150Hz, then filtering out that portion of the spectrum with a steep low-cut filter so it never reached the driver. The result was bass that sounded way deeper and cleaner than it had any right to, and speakers that could be turned up quite loud without suffering excessive cone excursion.

I don't know whether the implausibly clean-sounding small Bluetooth speakers that are ubiquitous today are doing much the same thing, but would not be at all surprised to find out that they are.

The human ear is nowhere near as sensitive to bass as to midrange, and quite a lot of what we hear as low bass is just our brains doing harmonics-based reconstruction even when the drivers can actually reproduce it accurately.
posted by flabdablet at 10:34 AM on March 28, 2023