www.audioscan.it

"Audioscan embodies the confluence of three stages of modern music technology: tape recording, electronic synthesis and processing, and personal-computer-based digital audio. It starts with recordings of street sounds, turns those into electronic instruments processed by electronic effects, and then creates the final composition in a digital audio workstation. Although all the steps (except final mixdown, which uses analog tape) were realized using modern digital technology, the result is very much in the spirit of the evolution of electronic music over the last half century.
The mid-20th-century advent of recording on magnetic tape enabled composers to record fragments of natural and industrial sounds and then manipulate those recordings in a process that became known as Music Concrête. Common tape manipulations included creating tape loops to continuously repeat a sound fragment, splicing disparate sounds together, reversing the tape to play sounds backward, juxtaposing two different-length loops of the same sound, varying playback speed, and so on.

The Audioscan project begins with recordings made at 1,580 streets and squares in Milano. Electronic-music techniques (on which more in a moment) were used to fashion playable musical instruments from the street recordings. In another step reminiscent of early tape composition, those instruments were then used to create melody, percussion, and background-ambience loops in the digital audio workstation Ableton Live. Tape technology makes its final appearance at the end of the process, when the finished composition is mastered on a TEAC A-3340 tape machine.
Although electro-mechanical processes occasionally appear in music as early as the late 19th century, the modern technology of electronic music derives from advances in electronics made during World War II and from the beginnings of digital computing. In the 1950s and 60s most electronic music was realized, at great time and expense, in commercial and academic laboratories. The processes developed during that period informed the design of the commercial electronic instruments that appeared over the next 30 years and of the personal-computer music software that evolved thereafter.

Building the 50 or so musical instruments used in Audioscan bears a direct lineage to early electronic composition. For one thing the process was labor intensive—each of the notes for each of the instruments was hand crafted from one of the street recordings. For another, the primary tools, Cycling 74 Max/MSP and Propellerhead Reason/Record, are reminiscent of early computer and modular-synthesis tools. Max, the precursor to Max/MSP, is named in honor of Bell Labs computer-music pioneer Max Mathews. It is a graphic programming language designed originally for controlling electronic instruments and then expanded (the MSP in Max/MSP) to also generate and manipulate audio. Propellerhead Reason/Record, although not modeled after a specific device, is a software implementation of a modular synthesizer augmented with the latest digital audio sampling and processing technology.

Audioscan Art Director Giorgio Sancristoforo built a custom processor in Max/MSP called Translator that he and collaborator Guiseppe Cordaro used to translate the street recordings into musical instruments. Translator is a resonant filter bank whose 46 filter bands are tuned at harmonic intervals, the same frequency relationship that characterizes the overtones of the strings on a stringed instrument such as a violin or piano. For each note on each instrument, one of the street recordings is processed through Translator, tuned to the pitch of the desired note.
Processing the street recordings in this way is one of the more ingenious facets of Audioscan. Because the street recordings capture not just random noise but include sonic events such as car horns, shouts, construction noises, and so on, Translator affects each of the recordings differently. That's what allows many distinctive instruments to be fashioned from the original recordings.

Playable electronic instruments were constructed from the Translator-processed sounds using the sampler modules in Reason. Each of the pitched instruments has a four-octave range. In addition, a variety of percussion sounds were created for use in Reason's Redrum drum sampler. The sampled instruments were then augmented by the many effects processors available in Reason.
Finally, the processed percussion and pitched instruments were used to generate music loops. Additional ambient sounds were fashioned from the street recordings using another Sancristoforo Max/MSP creation, Gleetchlab 3. The individual loops were then arranged to create the final composition in Ableton Live, bringing the process full-circle to the tape-loop paradigm.

Of course, the music is not in the technical details—the composition needs to stand on its own. Audioscan presents a broad sonic palette; you'll hear many different instruments ranging from tuned percussion to wind instruments to complex tonal washes. Some sections of Audioscan are floating and ethereal, whereas others are driving, rhythmic, and almost danceable. It is a testament to Audioscan's creators that, starting with field recordings of street noises, they've fashioned a highly varied hour of music that holds together as a cohesive unit. "

Len Sasso
April 2010