This image can be found at the Wikipedia Commons website, along with the audio file that was used to produce it.
Note that each pitch is represented, not by a single line, but a vertically aligned array of short horizontals, each representing a separate "overtone." This is what is known as the "spectrum" of the sound, and all sounds, musical or otherwise, have a spectrum.
What we see in the spectrogram is a reasonable image of what we actually hear, in strictly acoustic terms. But, obviously, this is not anything like what we hear psychologically, which for most of us will be a simple series of "notes." Contemplating the difference between a sonogram image of a musical performance and what it is we think we hear, can give us an idea of the degree of psycho-cultural processing we perform when we listen to music. Musical notes are, in fact, not simply acoustical but also semiotic, i.e., acoustic phenomena filtered through a symbolic system.
To clarify, I'll take the liberty of offering an extensive quote from my paper, Echoes of Our Forgotten Ancestors:
(to be continued . . . )
As linguist Roman Jakobson once noted, “[t]here is...exactly the same relationship between a musical value and its realizations as there is in language between a phoneme and the articulated sounds which represent this phoneme in speech” (1987: 456). In other words, a pitch class (or a time point class) and a vocable class (phoneme) operate in more or less the same way. In semiotic terms, music, like speech, possesses second articulation [i.e., the ability to break sounds into distinct phonemes]. But unlike speech it lacks first articulation (morphology, the basis for the signifier/signified relation).
A basic principle behind what we usually understand as music is in fact this field of tonal and/or rhythmic values which can produce pitch and/or time-point classes, i.e., “second articulation” (see Grauer 1993, 2000). This is not something to be taken for granted. Music is (traditionally) not made from raw sounds (with apologies to John Cage) but from sounds that are (with a nod to Claude Levi-Strauss) “cooked.”
To put it yet another way (with a further nod to Jacques Derrida), that famous “supplement,” music notation, was in some sense always already there, in the form of the tonal/metric “force fields” which give rise to the values, or notes, “inscribed” in music from the start. The existence of tuned pipes, either free or bundled into panpipes, is early evidence of this, as such pipes can already be regarded as a form of pitch notation, each pipe standing for a given note, the whole set for a particular scale.
What all this suggests is that early music may well have set the stage for language by providing a kind of laboratory for phonological and semantic experimentation. It is perhaps only a short step from the play of sung “nonsense” vocables and the construction of tuned pipes to the birth of signs. While one might need to rely on “native speakers” to puzzle out the phonology of a given verbal language, the “phonology” of music is, apparently, already given to us—i.e., we ourselves may already be “native speakers” of any and all (traditional) musical “dialects.” This could explain why we are able to enjoy, and also notate, so many different kinds of music (p. 43).
*By "working definition," I mean a definition that would seem to apply in the great majority of cases, but not necessarily all. Additionally, while it's been argued that a great many peoples have no word for what we call "music," it is also true that in almost all cases, there are words for singing and words for the playing of instruments. Thus, for the purposes of my "working definition," music can be understood in the context of either singing or playing or both together.