Flies as musicians (5) — Nature-Symphony 70 (improved version)

(Much improved version — but there's a further improved version at https://freesound.org/people/Philip_Goddard/sounds/762802/ .)
Nature-Symphony 70 (Flies as musicians 5: pentatonic and major, with pedal tone), which I'm also calling Rannoch Moor Moods, 3. — Much improved version, with greatly reduced noise, and more layers (= more notes). See explanation further below. This is now a 16-layer work — indeed technically 18-layer, because two of the layers are each an octave doubling. Both the latter layers are at the same pitch but are positioned differently, to give a fairly persistent impression of a deep and strong pedal note. The motif started off as a short descending pentatonic figure against that pedal note, and I added a major chord at the top of it. This gives us multiple points of tension, tugging at that major chord and its pentatonic undercarriage, with major sixth, tritone and major ninth (and its octave-wider version) to spike the mix's vibrancy (Phew!). The sum total of background noise from all those layers is now much reduced.

The Rannoch Moor connection is via music that haunted and pursued me many years ago, yes, out on Rannoch Moor (Scottish Highlands, UK), on my single-day solo mega-walks (some 25 miles total) from Corrour station to Ben Alder for an early summit lunch stop and then the long and challenging walk down and across much rough boggy moorland to Rannoch station in very good time for the evening train back to Fort William. Unlike my previous Rannoch Moor Moods, this uses a powerful and rather complex motif that didn't come to me till long past my Scottish Highlands outings, and which I didn't particularly associate with Rannoch Moor apart from its intense vibrancy based in a pedal note. However, using it with this particular recording does implant it in what I perceive as a landscape along the way on those crazy Rannoch Moor hikes of mine, but here with the brooding quality being transformed into something intensely vibrant, like some dynamo of our inner creative force.

The fly sound is not the continuous hum of mostly hoverflies that I've recorded before in the woods. Here we're out in the open on top of Cranbrook Down in warm to punishingly sweltery weather, high up to south of Fingle Bridge in the Teign Gorge, and with a far-reaching panorama around us. Here our aural attention is repeatedly drawn to the 'zing' of individual flies and bees coming from different directions, and sometimes circling close around the recorder. Here and there we get little flurries of linnet contact calls, also with the odd grasshopper sound. Generally the longer 'fly' tones, which give rise to chords, are from bumblebees, not true (two-winged) flies.

Unsatisfactory constraints on all the flies Nature-Symphonies are primarily that (a) the sequence of layer offset durations is fixed for the whole work, so every sound will repeat the same quasi-melodic / rhythmic pattern (no rhythmic variation apart from the flies' antics), and (b) such a large number of layers means a quite high sum total of background noise, even after my best endeavours at noise-reduction.
__________________________

Why a revised version?

This has been made possible by the noise-reduction breakthroughs that were introduced in my Nature-Symphonies 72(a/b) and 73. More about that in Techie stuff further below. You have only to compare the waveforms of this and the original version to see the dramatic difference.

Really the musical idea I wanted to use in Nature-Symphony 70 was too much for that layered format to handle well, because of the aforementioned summation of background noise from the large number of layers I'd need to use in order to fully represent that potent musical motif. In order to achieve the work at all I had to use 10 layers, with two of those being octave doublings, and that produced only a bare minimum essence of the motif. The summed background noise was a pestilence — a fog that made a multitude of very quiet musical events difficult to impossible to make out. I'm leaving the original N-S 70 in place (https://freesound.org/people/Philip_Goddard/sounds/754884/ ), just so you can listen to a bit to hear the difference between it and this version.

Now we have 16 layers, with the two octave doublings meaning that the background summation is effectively from 18 layers, and yet that summed background, although clearly audible, is much quieter! Indeed, if you listen to it particularly with headphones you'd realize that the improvement goes much further still, for the remaining background sound has a very different quality. Instead of being a 'fog', it's primarily a detailed, vibrantly 'alive' sort of sound, which doesn't significantly get in the way. Reason? — Because the original soundscape was so quiet there on top of Cranbrook Down (almost as though it were a still night) that my bringing up the recording afterwards to a usable level made the recorder's mic self-noise clearly audible, and that was the culprit giving us the fog.

In this version that fog, although still present, is now very low and significantly less than the real, mostly wanted, natural background sound from the actual soundscape, just amplified by my having the flies at a useful volume to make a musical work.

Another positive change is the greater sense of vibrancy brought in by the melodic extension, with its small rhythmic inflections, and also the greater dynamic range, which also makes small quiet details stand out more.
__________________________

I made the original recording in very warm and humid weather during a very extended lunch stop on 17 July 2024 on the top of Cranbrook Down (Cranbrook Castle, an ancient hill fort), high above the Teign Gorge, Drewsteignton, Devon, UK.

Advisory

Important! To get the best out of this, with its mass of detail, listen with high-grade headphones.

A 4-days-later mockup of this recording taking place.

Techie stuff:

Recorder was a Sony PCM-D100, with two nested custom Windcut furry windshields. It was placed on a Sirui carbon-fibre tripod, set at a low height to reduce wind disturbance.

Post-recording processing was to apply EQ in Audacity to correct for the muffling effect of the windshield.

_________________________

Noise reduction

This took a lot of trial-and-error and hair-tearing, to find an arrangement that not only produced sufficient noise-reduction but also didn't make the recording unusable from having too big a dynamic range.

Noise-reduction procedure was in three stages, all in Audacity (the WavePad noise-reduction functions are all too prone to producing artifacts or removing too much, so I never use them):

• the recently added OpenVino AI noise suppression module, using the deepfilternet2 model and attenuation limit set at 2 (not higher, to minimize dynamic range expansion). Amazing that it could be so helpful (with care!) for my purpose, considering that it's modelled for extraction of voices / vocals, not natural soundscapes!
• Audacity's standard noise reduction function (spectral subtraction, using a submitted sample of 'clean' and constant unwanted background noise), set to 6dB noise reduction, with frequency smoothing bands set to 3 instead of my previously normal 6 (the default). This way I got the reduction without any noticeable artifacts at all.
• Because the two stages so far have reduced amplitude and made the sound over-bright, I correct for that (amplify +6dB, EQ curve a 9db linear tilt (8K to 100Hz away from the treble).
• Repeated the OpenVino noise-suppression, again with attenuation limit set to 2, followed by another EQ tilt, this time just 3dB away from treble, and amplitude adjusted to get the best out of the bigger dynamic range.

Because the OpenVino processing increases dynamic range it wouldn't always be a good option, but here it does work well, though it was a pig for me to get right (multiple tries, and multiple times almost giving up on it)!
_________________________

Layer pitch shifts (semitones above / below original): +17, +12, +9, -7, +3, +5, 0, 0, -3, 0, -3, -5, -7, mix (-9+-21), same, -15.
Layer acoustics: all layers: middling foreground in cathedral.
Note that in this and all the other flies Nature-Symphonies I'm not changing speed of any layers; only pitch gets changed (using kHs Pitch Shifter).

One problem with my use of pitch-shifting without associated speed change is that the process is liable to cause distortions and artifacts in the sound, especially when raising pitch by large amounts. I assume that that is the cause of the rather harsh tone of layers 1 (especially) and 2, mainly in the few really loud peaks.

Please remember to give this recording a rating — Thank you!