A few years ago, I started a (non-musical) year-long project involving taking a photograph each day. The photographs were mostly taken with a Holga “toy camera” using 120 medium format film, scanned and posted at Dankerton. The process is a bit cumbersome, and while I took all the pictures, they never got processed and posted. Someday…

I like the structure of these year-long projects, and have a couple of music-based ones to do. Right now, I’m working on one inspired by John Cage’s tape pieces, particularly the Williams Mix.

I particularly like Cage’s penchant for what he called indeterminacy. Not purely randomized music, but elements of the composition are left to chance or choice, meaning that each performance is unique.

My piece is called AudioCard.



Each day for a year, record at least one short audio clip. The clips are organized into categories:

1. Human sounds
2. Animal sounds
3. Mechanical sounds (cars, trains, machines, etc.)
4. Natural sounds (wind, trees, water, etc.)
5. Electronic sounds (synthesized sounds – optional category)
6. Procedural sounds (sounds coming from unusual recording techniques, such as swinging microphone around, feedback, etc.)

These categories are suggestions, and another set of categories may be used.

At the end of the year, the clips are assembled together with the following transitions:

1. Cut
2. Cross Fade
3. Fade (fade out/fade in)
4. Silence

These can be thought of the four quadrants of a matrix describing two values – the length of the fades and the amount of overlap.

The score was designed such that a set of dice used for fantasy roll playing games (such as Dungeons & Dragons) is used for the randomization. A 6-sided die is used to decide the category, a 4-sided die is used to choose the transition type, and two ten-sided dice are used to determine clip starting points and duration (as percentages).

But it will probably be simpler to calculate the transition by randomizing the fade length and overlap values, and I will probably implement the program this way.

The piece is assembled in the following manner:

1. Choose category
2. Choose clip from that category
3. Choose percentage of clip to use
4. Choose starting point
5. Choose edit type
6. Choose transition duration
7. Repeat until desired overall duration is reached


I carry a Minidisc recorder around with me to record the daily clips. Each clip will be stored as individual WAVE files in a set of folders, one for each category. I am building a set of programs to do this kind of audio file manipulation, and they will be combined to execute the final project.

Since about Thanksgiving, I haven’t been so good about recording every day. Which is unfortunate, as my plan was to record a clip each day.

But, too many of my clips so far have been atmospheric (ambient noise, computer fans, air conditioners, etc.) and if I force myself to record something, it tends to be sounds of beautiful Downtown Burbank – or more specifically, the parking lot of the office as I leave work. If I’m lucky, I might catch the train passing, but otherwise, it’s the sounds of the freeway and maybe a few birds. And I don’t need more sounds of birds and cars.

A wider variety of sounds is definitely needed. For the second half of the project, I will have to spend more time arranging for noises as opposed to recording them as I find them.

Fortunately, I have a contact microphone, and I am not afraid to use it.

Post a Comment

Your email is never shared. Required fields are marked *