People with perfect pitch can easily reproduce a tune they've just heard on their instrument of choice.

Most trained musicians without perfect pitch can also do this, but they would have to experiment a bit first to find the correct notes.

However most people, without perfect pitch, and even without singing experience can easily sing or whistle a tune they've just heard which they haven't heard before. Or at least find it much less difficult to do so than with an instrument.

I would have thought that this is down to relative experience between the use of vocal chords and instruments. I am much more experienced in using my vocal chords to make sounds than any other instrument because I've used them for talking every day for most of my life.

However, there seems to be a contradiction in this logic because even people with no singing experience can reproduce a tune with their voice much more easily than a professional pianist can reproduce a tune by ear on the piano (without perfect pitch).

Why do I find it easier to reproduce tunes with my voice than with an instrument? Even if I am highly trained with that instrument? Why do I have to experiment with pitches on an instrument to reproduce a tune by ear, but I don't have to experiment with pitches when using my voice?

  • 10
    It is not true that most trained musicians without perfect pitch have to experiment to find the correct notes (depending on what you mean by "trained"). Experienced and skilled musicians can immediately play what they hear. Untrained singers do the same thing that untrained players of instruments to: they slide pitches around until they seem to work. Experience talking is not remotely close to experience discriminating and executing accurate pitches. Really, all of your premises seem wrong to me.
    – user39614
    Commented Mar 3, 2019 at 16:56
  • What instrument do you play? From my own experience I can tell you that my relative pitch was better because ai played a brass instrument as there you always have to mind the interval ... but then playing piano I improved my knowledge of harmony that helped me to derive the chords and tones I heard. Commented Mar 3, 2019 at 17:17
  • 2
    Perfect or absolute pitch has little to do with the capability of being able to reproduce a tune once it's been heard.Absolute pitch is being able to hear a sound and identify what pitch, or note it actually is. Second sentence is also not accurate. Many competent musicians who rely heavily on sight-reading to play may not be able to re-play something heard. Others, who have a 'good ear' will be able to reproduce a tune accurately.You are basically talking about a single note tune, as sing/whistle is all that can be.So anyone who knows their instrument as well as their voice will find it easy.
    – Tim
    Commented Mar 3, 2019 at 17:17
  • 1
    @AlbrechtHügli - that's as maybe. Whistling or singing produces only one note at a time, so no harmony in the question, which, to my mind, is asked using flawed premises.
    – Tim
    Commented Mar 3, 2019 at 17:19
  • 1
    I don't think people with absolute pitch would think in terms of a series of note names when trying to pick out a tune, even on a familiar instrument. I don't think it's what many folk do anyway. Absolute pitch is useful to recogise what note is being heard, not categorise a string of notes. As others say, relative pitch is far more useful here. And although those with a.p. probably have good r.p., it's the latter they'll use more.
    – Tim
    Commented Mar 3, 2019 at 18:19

9 Answers 9


Voice is an 'instrument' that many people learn in almost an ideal way. We start young, and practice with it very frequently - even speaking teaches you how to control the pitch of your voice. And because it's an instrument that everyone has, almost everyone also learns to sing in various environments, from singing rhymes in nursery school, to singing along in school assemblies or as parts of a sports crowd - not to mention those who learn in a slightly more formal way.

Voice is also an instrument we have relatively direct control over. We don't have to worry about things like finger positions - with a moderate amount of practice, most people develop a subconscious mental map of how to control their vocal cords to hit a certain note; it almost becomes like a "thought-controlled' instrument. Due to the immediacy of control, we can also adjust very quickly, so small mistakes can be corrected less noticeably than with other instruments.

Due to all these advantages that being 'built-in' to the body gives, it's true that many - perhaps most - people do have the ability to sing back a tune.

However, I think you're underestimating how good professional musicians are at reproducing tunes by ear. As per David Bowling's comment, I'd expect most professional instrumentalists to be able to play back tunes on their instrument by ear with a similar level of accuracy. It may be that they aren't able to do so with quite as little conscious thought as someone can when they are singing, but their experience with mapping heard notes to scales, and their scales to positions on the instrument, will make up for that.

Why do I have to experiment with pitches on an instrument to reproduce a tune by ear?

Probably simply because you haven't attained enough experience in reproducing tunes by ear on an instrument! To get good at this you probably need to learn from various angles. One angle would be doing some ear training exercises to help the mapping between the sound and the notion of which note of the scale; another exercise would be to practice scales and arpeggios on the instrument to reinforce the mapping between the concept of the note and the physical finger position. Practicing improvisation will also help to make your playing more instinctive. Even some quite 'passive' exercises - like noodling on the instrument while watching on TV - will help with this.

  • 1
    I would probably have to play a guiding note first, then I'd be able to play what I heard (relative pitch maybe?) Commented Jul 18, 2019 at 21:17

I would have thought that this is down to relative experience between the use of vocal chords and instruments. I am much more experienced in using my vocal chords to make sounds than any other instrument because I've used them for talking every day for most of my life.

As you say: even not trained people are able to reproduce a tune (singing or whistling). This must have become an unconditioned reflex by our language acquisition.

In opposition to your assumptions it can be aid that some musicians become a unison with their instrument that they can play every tune they hear without reflecting (right in the same way you are able to reproduce a tune.

To the most part of people the training of solfege and relative pitch is what they need to identify the tones of a melody.

The reflection of the names of the notes you sing or whistle will be necessary to play the right tones or keys on an instrument. It will take years of practice until this may become a conditioned reflex.


The way we produce sound with our voice is the simplest thing on earth: if you want to sing low, you open your throat as much as possible, and then gradually close it to produce higher sounds. So all you need to sing is the ability to recognize whether you are too low or too high.

On the other hand, when you play an instrument such as the piano, you can't adjust your note as fast as you'd do when singing. So if you are playing a note which is to low, you will have to make a guess on how much it is too low, and if you don't get it right away (which will probably happen) you'll have to do it again and again. This will take way more time than adjusting your throat takes.

I think this is also why when someone starts to whistle, he gets good really fast, even though he has not been practicing since being a kid.

If you play saxophone, clarinet, bassoon, trumpet, or anything that has an embouchure, try playing a song with only your embouchure, and you'll see it's not so hard (even though some do not have a huge range, like the saxophone's embouchure in which you can barely get an octave).

So yes, I'd say it's the way the voice works which makes it so easy to use.


Singing is monophonic - only one note at a time. Piano and guitar and drums and so forth can play lots of notes at once. each extra note will take another human to sing it.

Also, you can feel whether the note is right or not immediately when singing and quickly glissando to the proper note if you hit it wrong. hit the wrong note on an instrument and there's usually no saving yourself - it was just wrong and there's no hiding it.

But provided your instrument is monophonic, you'll eventually get to the point where you can "sing" with it.

With instruments that can play 2+ notes at once, you will always have to "just learn and practice it"... It's just gonna be harder because you're doing multiple things at once.

Singers have it easy (ok ok sorry sheesh).


Think about this from a computing perspective.

Voice is your built-in default output, optimized from the ground up for power-efficiency and instruction-cycle efficiency in the operating system and hardware.

Now consider replicating music on an instrument. An instrument is external hardware that was never optimized in the operating system or hardware it's being run on. In the computing world, using new peripherals either involves loading drivers (plug-ins at the operating system level) or using a program that comes with the peripherals (application-layer interface).

People with perfect pitch could be compared to having efficient operating-system level drivers for the new peripheral (instrument). Good, although slower than optimization in hardware.

Everyone else (with no perfect pitch) could be compared to using an application-layer interface (specific program) to control the peripheral (instrument). Slow and worst performance compared to OS-level drivers or hardware optimization.

Alternately consider a data-oriented perspective.

Audio input is fundamentally analog, and for purpose of discussion assume that the brain stores audio input in a simple analog format (similar to a WAV file on computer). Voice output natively accepts the same analog format, which means there's almost no processing overhead in echoing what you just heard.

Think of musical notes as a digital (logical) format - a sequence of note begin/end markers. To play an instrument, you have to run through the following algorithm. For purposes of simplicity, assume that you're only playing 1 instrument at a time.

instrument D = getCurrentInstrument().open()
P = NoteInputStream(AnalogAudioInputStream(input song), D)
while true:
  note N = P.nextNoteMarker()
  if N == null then exit
  else if N.isBegin then D.holdNoteAfterDelay(N, N.delay)
  else if N.isEnd then D.releaseNoteAfterDelay(N, N.delay)

Seems simple enough in pseudocode. But what's really going on? The complexity is in the API functions above.

getCurrentInstrument() is a function that returns a reference to the instrument in your hands. Relatively simple. The instrument.open() function is another thing entirely. This loads the drivers for the instrument - in other terms, you have to "recall" how to play the instrument. This includes mentally mapping logical notes to whatever you have to do on the instrument to produce the note (probably a reverse lookup table). If the instrument includes holding and releasing a lot of notes at the same time (like a piano), this also includes figuring out how to optimize the fingers for holding down a set of notes. Therefore, instrument.open() takes a lot of processing power and short-term memory (RAM).

AnalogAudioInputStream(audioInput) takes analog audio input and converts it into a "stream" of data.

NoteInputStream(analogInput, instrument) takes the above-mentioned analog audio stream and an instrument, extracting the notes for the specified instrument from the analog input. This step is where perfect pitch comes in, identifying the "main" instrument and mapping its frequency to logical notes in real-time.

Each note is represented as a key, octave, delay before processing, and begin/end flag that represents if the note represents a hold-down or stop-holding-down event.

Playing back the analog input over instrument: loop over the set of extracted nodes and hold/release on the instrument as dictated. Complexity of hold/release note operations depends widely on the instrument and obviously may fail if you run out of fingers or the input is ridiculous.


With your voice, you can kind of feel the notes you sing. However, on a piano or a guitar, you can't just simply feel which notes you want to play. You have to first of all know the notes on the instrument (i.e. which key on a piano is a C), what octave to play (i.e. C1, C2, C3, etc. on a piano), then which notes you have to play to reproduce the melody.

This isn't easy if you don't have perfect pitch because you can only guess which note to play.

However, if you have good relative pitch, which means you can guess notes based off of intervals, then you could better recreate the melody if you're given the starting note.

But if you don't have perfect pitch or good relative pitch, then this is a difficult process. With your voice, you can feel the notes you want to hit because it's your body and it's natural to you. But a piano or a guitar, or any other instrument, is not a part of your body and is therefore, not natural to you. So, you can only guess which notes to play.

Your relative pitch might not be well trained. You can do simple ear-training exercises, where you're given the starting note (or chord), and then a short melody is played. Then, you have to play the melody. The RCM (Royal Conservatory of Music) has good exercises to take advantage of. You can find good books called "Four Stars". You could also look up simple ear-training exercises on the internet.

Even if you're a master at the instrument, if your relative pitch isn't very accurate and you don't have perfect pitch, then that would explain why you're having a hard time.


This is a great question one that I've asked myself. And my theory is that it comes down to muscle memory. Since you were a kid you've been practicing whistling/humming so you can translate the tone that you have in your mind to your mouth right away. It's the same as singing. Most of us can sing in key with any song on the radio.

With an instrument, it's not so much about having a perfect pitch. It's just about not having that muscle memory developed so you fumble for the correct note to make that sound that you already have in your head. But the more you play, the more you practice all scales, and especially playing by ear, the more you start remembering what that note sounds like before you even place your fingers. And after awhile, with enough practice and time, you'll be able to "whistle" with your instrument. But even those with perfect pitch would need to train their fingers with their instrument to get that muscle memory to playback in realtime what they are hearing.


It looks as though no one knows why musical individuals can instantly and accurately sing or whistle any pitch from an instrument or another voice. This is true even if a group has gone a quarter tone flat. Also, a perfect match occurs even with a note from a tempered-scale tuned instrument, such as a piano. The issue there is that no set of notes can be "correct" for any key [to produce properly matched harmonics, the fifth of every chord would have to have a frequency ratio of 6/4 to the fundamental, and the third 5/4. In a tempered scale, every half step is exactly 2^(1/12)]. Although my third grade teacher confirmed that I had perfect pitch, I have avoided practicing it because that could not produce accurate pitches in an a capella group.


It might really be linked to a theory that at least some people developed - where a child at a young enough age that is trained in a particular way to get the appropriate brain (neural) pathways appropriately configured/set-up for absolute/perfect pitch feature, then the absolute/perfect pitch features may stay around with that person for a long time. Although, it is also known that some people lose absolute pitch ability in times of old age, or older age. But - ignoring this, it may well be linked to all or most humans developing pitch control of the voice from an early age - is because pretty much most of us had to develop voice language from a young age.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.