Classifying sounds

Classifying sounds

The GRM was synonymous with this interest in audio recording and the manipulation of recorded sound as the central material for music composition.

In 1952, Schaeffer published a treatise on the treatment of sound objects, classifying sound objects according to seven parameters:

• Mass: the sound spectrum (or timbre) of a sound object
• Dynamics: the loudness or presence of different components that make up a sound object
• Tone quality: the particular qualities or colour of the sound object
• Melodic profile: how the spectrum of the sound object changes over time
• Profile of mass: how spectral components of the sound object change over time
• Grain: the quality of any irregularities to the surface of the sound
• Pace: how amplitude dynamics of the sound object change over time

In applying these ideas we start to conceive of sounds as being composed of a number of component elements, all of which can be manipulated individually. For example, let’s listen to this door closing sound effect:

Here we can hear two distinct parts of the sound – the door creak and the door closing in the lock. But we can think of these two components as constructing a single sound object. We might describe this sound according to Schaeffer’s parameters as such:

• Mass: Door creak occupies the higher frequencies, and this shifts to a wider, lower range of the spectrum when the door closes
• Dynamics: Door creak is soft, while door closing is loud
• Tone quality: Door creak is thin, while the door closing is fuller
• Melodic profile: The door creak has a descending quality, while the door closing has a distinct, two pitch motif (a lower pitch followed by a higher pitch) as the door closes
• Profile of mass: The spectrum of the door creak gains frequencies as the door closes, the door closing seems to shift its frequencies as a block as the door is closed, rather than having elements added to or subtracted from it
• Grain: The sound appears to be very clear, absent of any additional sounds that seem unconnected to this sound object
• Pace: We can hear that the door creak subtly increases in dynamic volume, but that the sound object gets notably louder once the door closes

The TARTYP

These observations were expressed in Schaeffer’s Tableau Récapitulatif de la Typoloie, or the TARTYP

This was an attempt to visually represent all possible sound qualities. At first this can be confronting diagram to try to parse. Is it really possible to visually represent all possible sounds according to this diagram? While even Schaeffer conceded that assigning one sound object to exclusively one category might be difficult, what we can see here is a really intricate level of thought into the qualities of what defines a sound. This model has since been the inspiration for a number of taxonomic approaches to different kinds of sounds, with the TARTYP used to think about the qualities of sounds that define all kinds of electronic music, even hip hop!

Over to you

What are your impressions of the ideas of the GRM? How does your analysis of the door closing sound differ to the one we’ve presented here? How do you think the sound would change if one or more of these parameters could be interacted with by the composer? Answer in the comments below.

References

Holmes, T. (2008) Electronic and Experimental Music: Technology, Music and Culture. 3rd Edition. New York: Routledge.