The A-Zed of Audio
Can you taste the difference between a lemon and a strawberry? I hope so. If you, blindfolded, had a chunk of each fruit fed to you, how would you describe the difference? Certain attributes like texture and bitterness might be obvious but there are elements of the taste that'd be harder to describe. Yet your difficulty in finding the words wouldn’t mean you’d have any trouble distinguishing a lemon from a strawberry.
When it comes to sound we have a real challenge when we talk about timbre. Timbre isn't a Canadianized spelling of the word you yell when a tree is falling down or if you're Pitbull (feat. Ke$ha) wanting to make a night you won’t remember or be the one you won’t forget. The timbre we're talking about is pronounced TAM-burr and we know it comes from French but there isn’t a really clear translation for it.
Maybe I should clarify – if we have a working set of ears we know what timbre means, but we have a really tough time capturing it with words. You can get the gist of what timbre refers to by imagining a scenario. We'l make this scenario very scientific because we'll remove the humans from it, and imagine a completely perfect, breeze-less, climate controlled, and objective arrangement in which we have two instruments. To be thoroughly scientific we should imagine these to be two abstract instruments, but if this post starts reading too much like a lab report I'll probably lose you, so let' imagine we have a harmonica and a banjo. Both play the same G note, vibrating at 392 Hz, and both play at a volume of 72 decibels. With no difficulty at all, the observer (I guess we need to re-insert someone into our supremely scientific arrangement) can tell that the two instruments don’t sound the same.
But how can that be?!
If we lived in a two-dimensional universe in which pitch and loudness were the only two parameters by which sound could be described, it would be impossible to tell these two instruments apart. Fortunately, such impoverished universes are reserved for the imagination and for scientific thought experiments; our world is rich with sonic character beyond pitch and loudness.
Timbre describes all those attributes other than pitch and loudness which allow us to distinguish between two different sounds. If you feel this definition is somewhat vague, you're not alone. In an article published in Computer Music Journal, Stephen McAdams and Albert Bregman wrote timbre is a "waste-basket category for everything that cannot be labeled pitch or loudness". I suppose that's hard to deny but I have a hunch that if you gave Messrs. Bregman and McAdams a 500 mL glass containing 250 mL of water, they’d also complain it was half empty.
To their way of thinking, the fact that we knowingly use a single term to capture a huge range of attributes is a fault; in reality, our language is quite impoverished when it comes to describing sound. With this said, timbre isn't a complete mystery – there are a few quantifiable parameters that we can attribute to it.
Let's begin with envelope. Not the thing you stuff a letter into, in this case envelope refers to how suddenly a sound starts, how quickly a sound decreases to it sustained level, and then how long it takes to return to silence. A snare drum starts extremely quickly and returns to silence almost equally quickly. A bowed cello on the other hand has a gradual start to its sound, holds its sustained level fairly constantly, and then dies out gradually. To put it differently, the envelope is the shape of the sound. The reason that envelope and timbre are so closely related is that if we mess with the envelope, we really change the character of the sound. A piano has a distinct onset (or 'attack') to its sound because of the hammer striking the string, and if we remove that attack and play the piano recording, it will be much less piano-like.
Another significant contributor to timbre is the frequency content of a sound. If you read about overtones you know that beyond the fundamental frequency there are a whole cocktail of other frequencies that make up the sound we hear. Much like the taste of ketchup, for example. Tomatoes are ostensibly the "main" taste, but the taste of ketchup is made up of a whole host of ingredients including vinegar, reams of high fructose corn syrup, salt, onion powder, etc. If you began removing or replacing ingredients – or even tinkering with their relationships – you'd have a different tasting ketchup. Obvious, right?
Likewise with sound – if you alter the overtones present in a sound, you change the timbre. Timbre is just like flavour, but for our ears.
The reason this is all important is that timbre was once considered nothing more significant than a byproduct of an instrument's construction. A priceless Stradivarius violin has a pleasant timbre because its construction came much closer to perfection than any other violin. Composers in the Mozart and Beethoven days concerned themselves with timbre in an indirect way, which was their consideration of which instruments would play which parts. Yet their primary concerns were with melody, harmony, form, etc.
Today there are entire genres of music whose primary concern is the timbre of the sounds used, and compositions can exist with a single pitch sustained for 5 minutes containing interesting timbral action. This is an extreme example, but it works its way in less extreme form into the music we hear more regularly: if you listen to some of the most popular music without paying attention to timbre – that is, if you simply consider the melodies or the harmonies – they're hardly more advanced than nursery rhymes. Sometimes they really aren’t.
If your concerns aren't strictly musical, timbre is still important. Even in a voiceover, it' important to consider the timbre (or you could say tone) of your actor's voice. What story does it tell? This is a good question to ask and can shed some light on why only one of 100 talented voice actors will be selected for a gig to sell used cars over the radio, and a different one will be selected for the voice over of the new Godzilla movie trailer.
Whether you call it timbre, tone quality, colour, or something else entirely, this mysterious "wastebasket" to which we refer is familiar to all of us. If we have the ability to listen, we know what timbre is. The problem with calling this area of experience a wastebasket is we start to think there’s something wrong with the world that can’t be put into words. Well I've got news for you, Messrs. Bregman and McAdams: if you'e concerned with the world that can't be put into words, then a lot of music will find itself in the wastebasket. If that's the case, I'm going dumpster diving.
Jordan Mandel is a Digital Media Lab Instructor at the UWaterloo Stratford Campus, and writes for this blog regularly. His hobbies include lab rat psychology, dominoes, and silent letters. More of his work can be found at jordanmandel.com/blog, which is home to the award-winning satire rag, The Outa Times.