Hacker News Reader

I'm fascinated by this discussion because of how imprecise the terms "perfect," "absolute," "pseudo-absolute," etc. are. It's doubly interesting because we're describing things that happen in our own minds, with each of us having potentially different abilities and training and no common reference to what a "typical" person can do.

For me, sometimes, I can hear the opening note or chord of one song and it will immediately remind me of another unrelated song. When I check, they're almost always either in the same key or a semitone off. It happens most reliably when the timbres are similar. It's not a conscious thing, and the more I try to make it happen the less it ever does. Oddly enough though, I am terrible at identifying intervals. (Major third, perfect fifth, yeah, I know they're different and I can hear that, but ask me out of the blue which one was just played and I'll manage to get it wrong.)

Another thing I can do is play what is essentially a "mental recording" of a song I know well in my head. It's like if I had an iPod, and I felt some urge to listen to "La Grange" by ZZ Top, I can hear it in my head -- guitars, drums, vocals. I can skip around to whatever part interests me. I find that when I start tapping out the rhythms of what I'm recalling, or humming along with its melody, my recollection is pretty spot on when played against a real recording on speakers. Again maybe a semitone off at worst.

Both of those suggest to me that, somewhere in there, I do have an absolute pitch reference that I'm drawing from in an undisciplined way. If I sat down and practiced I bet I could refine it, but... why? Might be a neat party trick, but aside from that I don't really feel like I'm missing anything without it.

The meaning of perfect pitch is unambiguous. When a note is played someone with perfect pitch can identify that pitch (Ab, F, C#, etc) without a reference. Also known as absolute pitch.

If someone has relative pitch it means that they need to hear a reference pitch before they can identify the played pitch based on the interval between the two.

Perfect pitch may seem like a party trick for a casual musician, but if you're a dedicated musician the skill is an asset that allows you to execute a whole lot quicker. Occasionally it can be a curse, because you're more sensitive to out-of-tune notes.

> The meaning of perfect pitch is unambiguous. When a note is played someone with perfect pitch can identify that pitch (Ab, F, C#, etc) without a reference. Also known as absolute pitch.

How does that work with variations in concert pitch? As I understand it, although the most common standard is to tune the A above middle C to 440 Hz, it's not universal.

Are the differences too small to matter? Or does music played on instruments tuned to a different standard sound "off"? Or are the other standards sufficiently rare that most musicians will never encounter them?

Perfect pitch means that you can identify 440 Hz. Whether you call it A or slightly flat A is a matter of convention.

It's like saying water melts at 0 degrees Celcius vs 33 Fahrenheit. As long as you're talking about the same temperature, it doesn't matter.

The definition of relative pitch is unfortunately ambiguous, however--I've heard it defined as being able to pick out one note or several note, but having to rely on that reference note from memory to determine the pitch of other notes. I have A flat and E flat burned into my mind from playing Scott Joplin's Maple Leaf rag so many times as a teenager, so if I think hard I can figure out what any note is based on that reference. I can't quickly identify a note without doing that in my mind though, so I don't claim to have perfect pitch.

You have a slow perfect pitch

For some there is joy in the act itself even if the benefits are dubious at best. Just look at how much time we as a species have spent on mastering the rubik's cube.

Many of the responses to this comment imply that this is something fundamentally different than perfect pitch and give it some different name ("pseudo absolute pitch", "perfect relative pitch", etc.). While I concede that most people think of perfect pitch as something that is instilled in some way in early childhood rather than achievable through practice, I guess I would ask those commenters... why do you think the latter is fundamentally different, or is using some separate mechanism?

There are many skills which are much easier to instill in early childhood and are simply harder to master if approached in adulthood -- language learning, certain athletic skills, and more -- but we would never consider any of these impossible to achieve through study. Sure, maybe the maximum achievable skill level is less than what could have been possible if study began in early childhood, but we would not say that it is impossible for adults to achieve a level of mastery, or that those who gained a skill through serious practice must be using some separate mechanism than those who learned it in early life.

I contend that it is the same with absolute pitch. After all, there is not even a perfect level of absolute pitch mastery! In layman's terms, "perfect pitch" is usually understood to mean that a person can immediately name a pitch when played -- on a 12-tone western music scale. But some people people with perfect pitch have better precision than that and can estimate quarter tones, etc. If a note is played that's 20 cents sharper than Ab, and person #1 says "that's an Ab" while person #2 says "that's a note a touch sharper than Ab", most people consider neither statement to disqualify them from having absolute pitch. But there is a difference. Moreover, no person on Earth can name a pitch down to, say, a couple decimals of absolute frequency value. Doesn't this imply that the skill exists as achievable points on a spectrum, not as a flat binary?

In fact, there is an interesting historical anecdote about polish-american pianistic prodigy Josef Hofmann related to absolute pitch:

"Josef startled musicians by the accuracy of his ear. Once, at the Metropolitan Opera, he heard a tuning fork supposed to be at 440-A. Josef said it was a shade sharp, and it was."

This is incredibly impressive for two reasons:

- It shows absolute pitch is not necessarily limited solely to semitone identification

- A pure tone from a tuning fork doesn't have any of the characteristic overtones, timbre, etc. that you'd get from an instrument to help identify the sound

- It shows absolute pitch is not necessarily limited solely to semitone identification

Why would it be? There's nothing fundamental about a semitone. It's an arbitrary division that varies across different musical systems.

I'm not saying that it is but MOST people when they think of perfect pitch - they think in terms of note identification (e.g. semitones), A, A#, etc.

My point was more about how everyone's ability falls along a "Hz Range Level".

I imagine the gold standard of perfect pitch would be identifying exactly how many cents sharp or flat a given sound is--you could tune a piano by ear if you had that.

You can tune a piano by ear by listening to the beating. You need one note to be tuned according to a given reference but that reference doesn't have to be 440 hz

I personally do believe that there is something fundamentally different about perfect pitch perception, and think of it more like discussions around aphantasia or maybe synesthesia. I followed along with this guy's research years ago on his quest to understand and obtain perfect pitch [0] and was pretty obsessed myself. I had a number of friends with perfect pitch and would do random experiments and quiz them ("ooh, you're drunk, can you tell what note this is?" etc).

From that link:

> For the same reason, absolute listeners do not perceive pitch "height". They do not perceive pitches as "higher" or "lower" or physically "next to" each other. As musicians, of course, absolute listeners learn that, metaphorically, pitches are "higher" and "lower" than each other, because they can see these relationships on a page of sheet music. They also learn that, theoretically, "distance" between notes exists, because you can count the semitones that separate them, and you can see the "distance" between keys on an instrument. But, to an absolute listener, neither "height" nor "distance" has any direct perceptual reality.

> For example, when a non-absolute listener hears a guitar slide, we literally hear something moving down. But an absolute listener's experience is nearer to the color-changing rectangle above. They hear a series of discrete pitches, changing—not moving—from one to another. Although they know the sound has "descended" from their knowledge of the musical scale, the sound does not give them a literal experience of downward movement as it does to non-absolute listeners.

The link has a gif of a box moving through the color spectrum which helps understand the point.

[0] http://www.aruffo.com/eartraining/

> While I concede that most people think of perfect pitch as something that is instilled in some way in early childhood rather than achievable through practice, I guess I would ask those commenters... why do you think the latter is fundamentally different, or is using some separate mechanism?

As soon as having learned some relative pitch, it becomes very difficult to train absolute pitch. One single note a day. For the second note the brain will switch to relative mode, and there's no progress on absolute. Forcing the ear/brain to run in absolute mode seems possible, but is difficult consistently enough to practice absolute.

Source: Experimenting with my own ability (or lack thereof).

> A simple way to do this is to make playlists for a given key, which helps reinforce the sense of shared tonality across songs.

This is gold. I similarly maintain lists of melodies that prominently feature certain intervals.

I also have a repertoire of songs whose keys I've memorized to function as anchors, then using relative pitch to gauge whatever note I'm trying to discern.

I'd like to skip this relative pitch translation and jump straight to recognizing each note directly, but I have gaps in coverage. I'm happy you brought up "Comfortably Numb" since it's also burned into my neurons - just now I thought about it for a moment, went to the keyboard and hit B exactly! Also love "Hook" and will never be able to forget that harmonica intro.

Some of my songs:

  * Vampire Weekend - "Mansard Roof" in F major
  * Radiohead - "Everything in its right place" first chord: C major
  * Queen - "Bohemian Rhapsody" piano part in Bb major

I'd be happy to see any "songs all in the same key" type playlists if anyone wants to share one

I'm a programmer interested in music visualization and it would be handy to have a few of these

Not a playlist, but I've always hit up Song Key Finder[1] when trying to figure out what songs might fit together for mashup-type purposes. They rank the songs by popularity within their user community.

[1]: https://www.songkeyfinder.com/

Not a playlist per se, but https://www.hooktheory.com/cheat-sheet/key-popularity

Has popular songs for each key and mode

This is the way. Most people don't have 'absolute perfect pitch' but, almost anyone can train themselves to have relative pitch, which can mostly give the same result. That is when you are given a good starting point to refer from, you can then work out where something is relative to that starting point.

That's why tuning forks and pitch pipes exist. You need a source of truth to ground things in, and then you can train yourself to hear the relative distance between notes from there.

Its worth noting that adults can learn something called pseudo absolute pitch, which sounds similar to what you're describing. The difference (which you're also describing) is that it requires reinforcement, and is a bit less accurate than having perfect pitch

It seems like a "bit less accurate" is a misnomer here. If you can successfully employ this technique, and use calibrated test equipment to measure it, you have in fact demonstrated your accuracy in recalling the pitch of notes, also known as perfect pitch. The means of achieving this recall do not in any way affect the accuracy of the results, because the results exclusively speak for themselves.

What you are describing is known as "perfect relative pitch." It's quite a feat to acquire it -- takes many years of practice.

My understanding of true, perfect pitch is something a bit more innate. Individual with perfect pitch do not need to rely on any other techniques.

I've known people with both types. If anything, the practical differences are not significant enough to truly matter from my observations. Many of the folks with true, perfect pitch have told me that it is a curse more than a blessing. These people tend to hear notes even when they do not want to e.g., toilets flushing in an out of tune C, the pitches in people's voices, the pitches of jet engines, CPU coil whining in D#, etc..

When a lot of modern life is constant noise and out of tune, it can be somewhat distressing for people with perfect pitch, or at least, so they have told me. I can't imagine turning off a song on the radio solely because it's a few cents out of tune, but hey, I knew people that would.

I've got far from perfect pitch, but I can relate. In high school (mid 90's) I took up guitar and figured out that I could use (the memory of) the intro to "November Rain" by Guns N' Roses as a way to mentally get an "E" in my head to tune the lowest string and go from there.

I'd heard the song so many times that I found I could recall the low, then high E at the start and that's really all you need to get the rest in tune.

Uhh.... What? There's no guarantee that the song you hear ringing in your head after a certain time has passed is going to be in the original key, in fact there's a very good chance (depending on your range) that you'll end up subtly transposing it to a different key when you try singing it in the shower apropos of nothing.

This is by no means a guaranteed method to learn perfect pitch.

I believe my claim of having perfect pitch stands on solid ground, and consists largely of what OP is describing. That's because I can demonstrate it. I wouldn't go so far as to say that the described method is a reasonable way to _acquire_ perfect pitch, but I do contend that what is described is a reasonable method of demonstrating and explaining one's perfect pitch. If you do have perfect pitch, that is reflected in the fact that you don't transpose the music you hear on in your head.

This idea of recalling songs is an easy way to explain perfect pitch to others in a way they can easily try for themselves. If you can sing the first note to a song, and then afterwards you actually listen to that song, and you are actually in tune, then you have perfect pitch. That's a far cry from being able to name the note, or rattle off the notes in a chord, but in my view the mental recall is the true innate essence that undergirds every other component of the umbrella of perfect pitch.

Yeah I would agree that with proper practice (assuming you have a "locked in" song for every semitone in a scale) - that this could probably be extrapolated to being able to hear a random note on an instrument and then naming that note, and vice versa singing a given note on command.

An even easier method, which I was taught is interval training. The jaws theme is a minor second. Do re mi starts with a major second, and so on. Interval training is easier, imo, and would then allow you to memorize fewer songs with absolute notes.

Yeah agreed. I did interval training as part of a sight singing class a while back.

The ability is common: https://www.universityofcalifornia.edu/news/singing-memory-u...

Interesting but prior studies in a more controlled environment show that the ability to recall a song in the correct key is only about 15%.

From the article: "Psychologists wanted to study “earworms,” the types of songs that get stuck in your head and play automatically on a loop. So they asked people to sing out any earworms they were experiencing and record them on their phones when prompted at random times throughout the day. When researchers analyzed the recordings, they found that a remarkable proportion of them perfectly matched the pitch of the original songs they were based upon.

More specifically, 44.7 percent of recordings had a pitch error of 0 semitones, and 68.9 percent were accurate within 1 semitone of the original song. These findings were recently published in the journal Attention, Perception, & Psychophysics."

This study seems extraordinarily suspect because it's essentially relying on good faith that the person recalling this song hasn't just literally heard this song on the radio a few hours before that or in the morning, and now it's an "earworm."

EDIT: The actual published paper makes it more clear how this study was done. Participants (N = 30, convenience sample) were surveyed at random times over a week and asked to produce a sung recording of any music they were experiencing in their heads. This is a little bit more reliable, but still hard to say how long this pitch perfect recall lasted from the last time that they actually heard the song.

Oh wow, I’ve always thought this would be a good way to do it but I’ve never bothered to actually look up the key of the songs I can “hear” in my head. If the “Hello” of Comfortably Numb is an F#, that’s all I need :)

wow you just made me realize I have some degree of perfect pitch. I can sing moonlight sonata's G#C#F# on command.

Can you share the playlists?