Sound 2

The external sound is low frequency, so I thought that a relatively high-frequency voice like that of a female would contrast and stand out acoustically from the background sound, particularly when the latter increased in volume. An example of this would be the ‘everlasting’ in the Symposium 2 video

Why a single female voice?

I feel that it would be in keeping for me to record this as a female voice but using my own male voice plays with an ambivalence of timbre achieved by digital means. On a more technical level, I want the soft female voice to cut through the noise coming from the subwoofers.

Another alternative would be to give Logos manifold voices. The thing is to experiment with different combinations. It could be argued that a single voice would imply an individuality fostering a direct connection. On the other hand, many voices could signify a wider community/society, humanity even but lose on the way some of its intimacy. At least I have the option for either.

Scripting Decisions

The scripting is another point of decision making. For now, I have opted for short phrases or words to beckon the person closer. These are perhaps more distinguishable from the surrounding environment and convey meaning more clearly than having to stand there listening to a long monologue. This is an area that offers a great deal of flexibility. Again, I lean towards a female voice, but it is not a deal-breaker. I also think that the higher frequencies contained in sibilance or whispering are directional, pointing directly to the sculpture as their source and cut through the omnidirectional sounds from the subwoofer low-frequency sounds.

The origins of the sounds

All the sounds in Logos are made using my voice. Firstly using my voice allows for a broad field of experimentation on altering sound which can be applied to other field and synthesised recording. Secondly, it is consistent with my idea of transformation. (Each step is like a mutation: I work heuristically until something works. This, however, differs from an evolutionary system in that it is wholely purpose-led. In the future, it would be interesting to look at generating sounds stochastically which then fall into niches, predetermined or not, that allows them to be saved and used. The criterion for choice could be many, aesthetic, content, fit for a purpose which the process is blind to. This reminds me of Latham’s work with computer-generated ‘life’.)

Altering sounds

The following samples were made by recording my voice, which is a baritone, and applying a number of filters in audacity to achieve the sought after effect, these were:

• change pitch
• reverb
• echo
• distortion hard limiter
• Normaliser

Having applied the echo, some phonemes repeat in an unnatural fashion, a little like a stutter. I delete these out at points of zero amplitude.

The permutations are almost endless. The results can change meaning, context and response.

Limitations to these experiments what to do next

Without the completed sculpture, I am unable to experiment and test how its internal chamber might affect sound. And therefore these samples act only as an indication of how I would imagine the sounds to be.

The words, ‘come nearer, come near, don’t touch me’ are deliberately intructive. They are chosen for their relevant to still sculpture. They could change, evolve. With the use of various sensors, voices could be activated in response to different viewer actions or positions. For example, code could be applied to one voice that was activated to stepping away with words such as, ‘come back, don’t go, don’t leave’.

Thinking this through, this acoustic device could be applied to any still sculpture. Therein lies an interesting piece of research. From monolith to complex installation to screen, the conceptual device gives the work an agency that still, silent or ‘scripted narrative’ often do not have. It addresses the viewer directly and might give the semblance of a limited intersubjectivity even if the visitor knows it for what it is. This is an interesting exploration in itself which also leads to the use of AI. Now that could be an interesting collaboration with someone.

Simultation

I have composed a sample simulation with the low sound destined for the subwoofer and the sibilant voice for the sculpture’s interior. It is a short, programmatic piece where the viewer is beckoned to approach, then move away a little before leaving altogether. The subwoofers can be head reacting to her proximity while the voice from within the sculpture speaks.

To listen to the full effect please do so on ear or headphones. This last sample, if listened to on laptop or tablet speaker, demonstrates the difference between low and higher frequencies. The speakers on such devices are unable to deliver the low vibrations. This is what makes it possible to combine the two soundtrack types.

Why use use Audacity?

There are many DAWs and programmes for altering sound recordings and generating them from scratch: Cakewalk, Audition, Reaper, Ableton, etc. Each one offers a particular advantage. Audacity is the first programme I go to for two reasons: familiarity and directness. It was one of the first programmes I used, it is free, open-source and very flexible. What you can do with most other sound software you can do with Audacity. However, its disadvantages are various. It works with a destructive process, that is to say, you cannot go back to previous states once saved so you have to save versions as you go along. This makes for an untidy and complex file system. Another disadvantage is that the plugins available are not always very sophisticated which means you have to work harder to achieve a given aim. This is not altogether bad in itself, it makes me think about what I am doing. However, the finness of parameters is not always there. Although other software offers other things that Audacity cannot give, it is my first port of call whenever I work with a sound recording.

Binaural?

Although I have explored working binaurally and used it in an instance in the symposium video, there is not really a space for it in this installation: it could prove distracting. However, I look forward to employing this in purely acoustic work.

An ongoing experiment in low frequency for intergration with Logos.

My aim is to produce a low-frequency sound that does not interfere with other sounds by reducing the detail of content but keeping it rich in texture.

A prime concern is to create a physical response without discomfort. It is very easy to produce nausea with low frequencies as it can disturb the balance organs in the ears. This ethical consideration is part of the installation’s risk assessment

The track is twenty three-and-a-half minutes long and looped. The following are brief samples taken from the same area of the track demonstrating some of the steps taken to reach the final mix.

I started with a basic sound generated with the voice projected into one of the sculpture’s (Logos) forms. The file was stretched, lowered the pitch and reverbed trying not to lose texture.

A lot of parameters and steps are involved in this process and experimenting with respective levels of gain is an important aspect of the work.

I then produced a duplicate track and reduced the pitch to around 68 Hz.

I also applied a 3 Hz shift between the left and the right channels to create an almost imperceptible pulse.

I tried one at 20Hz but the level of discomfort was far too noticeable. It also did not create the relatively smooth texture I was looking for.

The plan is to send the left and right channels each to one of two subwoofers so that they subtlely interact with one another.

The idea is for the sound to increase as the viewer approaches the work, not too much so as to create alarm or discomfort, but enough to act as a form of membrane barrier working against the softer sounds emanating from the sculpture itself.

I mixed duplicate tracks, balanced the levels of the original so that they are barely audible but add texture to the sound.

I compressed the files quite heavily to even out natural fluctuation in the voice and render smooth the proximity sensor-activated volume changes. The tracks were normalised to -3 dB

This final sample simulates a visitor approaching the sculpture, moving away a little return again before leaving the sensor field.

Finally, the track is put onto a media player hooked up to the Arduino coupled with the sensor. I explored doing this during Unit 1 last year.

The theory behind making a binaural microphone system is pretty straight forward. Geeks might argue over which is the best set, the £6000 one or the latest in ear gear. If binaural is going to be listened to over the internet using some average head or earphones, then I imagine that the following plan would work well enough.

The idea of doing this came when looking at the Whatsapp conversation in the ‘fineartdigital’ group. Friederike has posted some binaural pop tracks which were so effective that it really did sound as thought the music were coming from outside the head and not in the ear.

The effect is not only due to the proximity of the sound being recorded nearer one microphone than its stereo other. This panning alone can be done in a sound editor, but it results in a rather flat movement from one side to the other.

The art of creating a spatial effect is somewhat more complicated than simply panning. For it to be as true to life as possible, the gain (or volume) of the sound does not alter all that much as in panning. I noticed as much in the Friederike’s music track. This made me think as to what else could be causing this effect.

If one ear is closer to the source of the sound than the other, and there is little or no connection between the two ears on account of a dense head being between the two, then the only other reason for the spatial effect must be the time it takes for the sound to reach either ear.

The nearest ear to the source will receive the sound a fraction of a second before the other ear. Perhaps as little as 1 millisecond. This delay then creates the sense of space as the brain triangulates the location of the source with respect to the head. And as the source moves further away from the microphones, the gain also diminishes and gives the sensation of being placed further away, or further outside the head.

How clear this effect is probably depends not only on the arrangement of the microphones but also their quality. But as I mentioned above, I think a pair of average lavelier microphones would do.

All this could be achieved in an editor or digital audio workstation or DAW. However, the amount of work in graduating not only the panning but the delay would make this a very complex task. Perhaps there is some software out there that can achieve the same thing but I suspect that most if not all such recordings are produced using a binaural setup rather than in postproduction. Post production could be used, however, to enhance the effect.

I could build a binaural recording setup with two lavelier microphones, the Zoom H5 recorder as a preamp and some other bits and bobs. These would include: a dense material with which to simulate the width and mass of the head to isolate one mic from the other. The mics would have to be placed one head width apart. And I could fashion something resembling ears to act as actual ears and simulate the directionality of the listening organ.

I know of set ups which use a polystyrene head. I think there are better materials that can be used. Polystyrene is nowhere near as dense as a head and beside, the whole thing become unwieldy and silly looking. There is a set that can be bought which consists of a rectangular material one head width apart with very large, what appear to be, silicon ears. It is very expensive so I imagine that what maters is the distance apart and the density of the intervening mass to create the right amount of isolation.

The microphones have to be mono, which laveliers usual are and most importantly omnidirectional. Apparently electrotet mics are the ones to use. These are small so they need a preamp hence the Zoom. I could buy the mic capsules, but this would involve me in having to rig circuits with resitors and jackets and what have you. It really is not worth the trouble: it would only save a few pence and I would end up spending valuable time building the mics from scratch.

I could experiment with the ear configurations and make say, jackal’s ears – what the divine Anubis heard in the tomb!

https://en.wikipedia.org/wiki/Binaural_recording