Woodwinds with electronics

In a recent recital I performed my own arrangement of Ravel’s Boléro for multiple woodwinds soloist using electronics, with piano and snare drum. I used electronics to try to approximate some of Ravel’s harmonies (and timbres), and used what in my mind are three different techniques, which I’ll try to outline here.

In performance, I used the BOSS GT-1000CORE guitar multi-effects unit to do most of the heavy lifting. I did find that it had difficulty tracking my flute playing (though, surprisingly, it did better with piccolo), so I used an Electro-Harmonix Pitch Fork + to assist with that instrument in particular. I also used a BOSS SY-200 to try to create some non-woodwind tone colors. There are plenty of other equipment options that can achieve similar effects, but you’re on your own to read the manuals.

All of this was done with a microphone rather than pickups, which was less complicated for quick instrument switches, but did make it difficult to get relatively isolated woodwind sounds into the electronics, which ultimately caused problems with the audibility of some of the electronic sounds.

The warts-and-all live performance is available on my YouTube channel, but I’ll provide some clearer, isolated examples here. For rehearsal numbers in the orchestral score, I’m referring to the Durand Edition on IMSLP.

Technique 1: parallel intervals

At rehearsal mark 8 in the score, horn and celeste play the A theme in octaves, with two piccolos playing in parallel a perfect fifth and a major tenth above the celeste’s highest octave, perhaps in imitation of a pipe organ’s mixture stop.

To achieve this harmony with electronics, I played the upper piccolo part “live,” and routed the piccolo’s sound into the GT-1000CORE where I split it into two separate signal paths. One got transposed down using a digital pitch shifter to create the second piccolo part. The other got transposed down to the melody pitch and split into octaves, then routed through the SY-200 to turn the sound into something vaguely celeste-like.

Since the intervals are strictly parallel, this is a pretty straightforward use of pitch shifting: whatever note I play on the piccolo gets transposed to the specified intervals.

Technique 2: smart harmonization

At rehearsal mark 16, a thickly-orchestrated ensemble of woodwinds, brass, and strings plays the A theme in harmony. I opted to play this portion on soprano saxophone, thickened and harmonized with a synthesized string section.

Since the harmony in this section is largely diatonic, I used the GT-1000CORE’s smart harmonizer. I added voices a diatonic fourth and diatonic sixth below in the key of G (like a first-inversion triad), which tracks with the notes in the first part of the theme. But there’s a moment in the first part that uses F-natural instead of F-sharp, and the second part of the melody uses F-naturals exclusively, so I used the unit’s footswitches to change to the key of C major as needed. I routed all of this through the SY-200 to change the three soprano saxophones into a string section sound, with the “live” soprano remaining audible in the room.

For the key switching, I set one footswitch as a “momentary” switch, so it changes the key just while I’m pressing it, and another as a “toggle” switch, so I can press and release it and the key remains changed. This gives me some helpful options for live performance.

Technique 3: smart harmonization with custom scales

The smart harmonizer works well out of the box as long as you want to use notes of a major scale (or mode thereof), but at rehearsal mark 15 Ravel’s harmonization is more complicated than that. Luckily, the GT-1000CORE supports smart harmonization with custom “scales.” What this really means is that I can tell the unit that any time I play a certain pitch, it should add one or more pitches that I can specify arbitrarily. I can add whatever pitches I like to each note of the chromatic scale.

I chose to play this section on clarinet, using the electronics to turn it into a 3-part clarinet section. During the first phrase, the melody pitches are harmonized in a consistent way: every time there’s a melody concert B-flat it’s harmonized with a G and an E, every time there’s a melody C it’s harmonized with an A and an F, every time there’s a melody E it’s harmonized with a C and a G, and so forth. I can just tell the effects unit which harmony notes to add to each melody note.

But things change in the first half of the second phrase: melody B-flat is now harmonized with G and D, and C is now harmonized with A and E. To accommodate this I have to create a second custom “scale,” and use a footswitch to activate it at the right time. To finish the second phrase requires a third scale, engaged with another footswitch.

Because of the flexibility of the custom scale system, I can recreate harmonies that use a variety of intervals. With a little analysis I can figure out where the scale changes need to be (basically anywhere a given melody pitch is harmonized in a new way).

Additional thoughts

There are some limitations to using pitch shifters and harmonizers, depending on your equipment. Each virtual pitch shifter and harmonizer in the GT-1000CORE can only add two voices, though by (virtually) splitting the audio signal into multiple paths and passing each through its own shifter/harmonizer I can build thicker chords.

When trying to reproduce specific harmonies written by a composer, there may be some decisions to make to balance accuracy with practicality. Serendipitously, most of Ravel’s harmony translated fairly easily to the effects unit’s capabilities. But there were a few spots where I decided that certain chord voicings were close enough, and that I didn’t need to complicate things with one more custom scale plus the corresponding onstage footwork.

As I’ve mentioned previously, I’m more interested in electronic effects that give my woodwinds new capabilities, like polyphony, than in just adding some distortion or echo (though those are also fun). Enjoy!