I was wondering about the relative frequency of certain topics that intrigue me in regard to contemporary, technology-oriented sound studies. For a quick glimpse, first I charted “machine listening” in Google Trends, then adding “audio surveillance” and, for a broader swath, “audio deepfake.” The three terms were almost identical in the narrow band of popularity they populated for the past few years. I sought to expand the subject matter with a fourth item, something related to artificial intelligence.
Needless to say, we’re in the midst of AI Summer. These days “AI + [anything]” — following the rise in chatter around DALL-E 2, Midjourney, Stable Diffusion, ChatGPT, DeepMind, Bard, and OpenAI, among other projects — is going to be more popular now than it was even six or seven months ago.
Still, when I selected a term, I was fascinated to see how much larger this one line item was than the others. To round out the set, I added what I imagined would be the more active “voice ai,” yielding the outlier green bar in the chart below. The results marked a significant shift:
What’s going on isn’t merely that “voice AI” is greater than the other three lines combined. It’s that “voice AI” renders any internal divergence among those three to a flattened data nothing-burger. To a degree, this divide might signal that a single term has become a catchall for all manner of subjects, fields, and anxieties in the popular imagination. Alternately, it could also mean that whatever energy might be spent more in more different ways is now focused, laser tight, on where the SEO money is. What would be great would be to see a wider variety of terms gain traction as time passes.
This went out today as a thank you to paid This Week in Sound subscribers: an annotated playlist of recommended music (a new Aphex Twin cover from Simon Farintosh, a science fiction audio backdrop from Star Trek, and a noise album from Tim Olive).
These sound-studies highlights of the week originally appeared in the June 6, 2023, issue of the Disquiet.com weekly email newsletter, This Week in Sound. This Week in Sound is the best way I’ve found to process material I come across. Your support provides resources and encouragement. Most issues are free. A weekly annotated ambient-music mixtape is for paid subscribers. Thanks.
▰ FLIGHT CLUB: “[A] growing body of research is showing that the affinity human musicians feel toward birdsong has a strong scientific basis. Scientists are understanding more about avian species’ ability to learn, interpret and produce songs much like our own.” One researcher (Hollis Taylor, a violinist and an ornithologist) “has observed what appear to be warm-up sessions, rehearsals and singing contests. Other than humans, there’s only a ‘small club’ of species with an observed capacity to learn songs and vocal patterns.” And mysteries remain: “If birdsong’s main purpose in some species is for males to attract females, then why do some females also sing? ‘Female song actually arose very early in songbird evolution,’ he said. ‘In species where females don’t sing, it’s because they’ve lost the ability to sing rather than it being gained.’ This indicates that it may have once been evolutionarily beneficial for females to sing — and scientists can’t say why.” (Read at nytimes.com with a gift link.)
▰ HELL’S BELLS: “Yes, the harvested audio will be imported onto a computer and deepened, sculpted, flayed, and spliced until it fits the unforgiving grim-dark horrors of Sanctuary, but Blizzard still takes a distinctly classical approach to the aural aesthetics of Diablo IV, one that resembles the practical Hollywood filmmaking of the 1950s and ’60s. The marauding demons are programmed with dangling bike chains, molten candle wax, and crushed fruits and vegetables, all of which is captured tangibly, without resorting to the freeware clips bobbing around the internet.” All about the sounds of the new video game.
▰ AMPED UP: “He describes Sonic Check as ‘rapid-development tool’ that uses machine learning AI ‘trained with real market research’ to give the user a ‘measurement’ of a sound. It looks at how consumers responded to similar sounds in the past and provides a prediction of how a sound will be received by consumers. … Once users upload a sound, the AI analyses its performance in relation to ‘brand fit’, memorability, and ‘authenticity’.” —Abbey Bamford interviews sonic branding agency Amp CEO Michele Arnese about Sonic Hub, “which seeks to simplify the sonic branding process for designers and brands using four different AI technologies.” Sonic Check is one of its tools. The others are Sonic Radar (which provides “insights on a brand’s use of sounds across its digital channels, using music AI tagging technology ‘trained by experts to categorise music’”) and Sonic Space (which “uses generative AI to create new music out of existing music, acting as a ‘sonic repository’”). More at ampsoundbranding.com/sonic-hub.
▰ QUICK NOTES: ▰ This Is What It Sounds Like: The Shriek of the Week is the “stock dove,” which, we’re told, sounds like a normal dove’s “coo” — but “in reverse” (“Imagine the bird is scratching a record like a DJ, swiping it backwards and forwards”). ▰ Drop the Mic: Micah Loewinger of On the Media speaks to Dan Charnas, author of Dilla Time: The Life and Afterlife of J Dilla, the Hip-Hop Producer Who Reinvented Rhythm, about “how music copyright law suppresses the artistic voices of hip hop producers.” (Thanks, Rich Pettus!) ▰ Light Brigade:Behind the scenes with the Imagineers who developed the sound for the TRON Lightcycle / Run ride at Walt Disney World. ▰ Songs About Buildings: Learn all about new Finnish acoustic standards for “acoustic environments in buildings.” ▰ Grand Scale: Composer-developer Giorgio Sancristoforo’s new software is a synthesizer with 1,000 oscillators. ▰ Sized Up: How MassiveMusic Berlin updated the sonic logo of the German bank Sparkasse. ▰ Voice Over: The voices of Shaquille O’Neal, Melissa McCarthy, and Samuel L. Jackson are leaving Amazon Alexa.
▰ APPLE DROP: There was a lot of Apple news this week, and in it quite a bit of audio, such that it gets its own section this week:Coming in Apple tvOS 17 is “Enhance Dialogue” (“which lets users more clearly hear what is being said over effects, action, and music in a move or a TV show” — thanks, Bruce Levenstein!) ▰ The new VR goggles from Apple involve “audio ray tracing”; here’s a primer: wepc.com (“It would make it seem like … sounds are coming from your room in a particular place.) ▰ “There’s a new Adaptive Audio feature for the AirPods that combines Transparency and Active Noise Cancellation to dynamically match the conditions of the environment that you’re in.” ▰ USB-C microphones to be supported by iPads. ▰ Apple Music adds crossfading, among other features. ▰ “Users can now simply say ‘Siri’ instead of ‘Hey Siri,’ and Siri will understand follow-up commands that do not include the trigger word.” ▰ “Personal Voice is designed to allow you to use artificial intelligence to create a replica of your voice.” ▰ Added to iMessage: “automatic transcriptions for voice messages.” ▰ FaceTime will have “Live Voicemail with voice-to-text transcription before answering; transcription is handled on device.” ▰ “AirPlaycan learn how and when you listen to certain content, for example by displaying a nearby AirPlay-supporting speaker to select depending which room you’re in.”
From Orkney Islands, Scotland: discarding nothing, living remotely, working with Pure Data
/ By Marc Weidenbaum
This Junto Profile is part of an ongoing series of short Q&As that provide some background on various individuals who participate regularly in the online Disquiet Junto music community.
What’s your name? Ray Cobley (Raymond C.A. Cobley)
Where are you located? Stromness, Orkney Islands, Scotland. I have lived in Orkney for most of my adult life, including around 20 years on the small island of Sanday. Before that I moved around a lot, living in various places in the UK from childhood on, also spending some years in Germany, Austria, and Spain. In my youth I studied languages, gaining a master’s degree in German literature and a diploma in Spanish. I have had various occupations in teaching, agriculture, and retail, mainly on a self-employed basis. These days I am retired from paid work.
What is your musical activity? I kept the previous section brief because it has little direct bearing on my current activities. Throughout all the various stages in my life, music has been a constant and continuous thread. I had a solid basic grounding in music theory and have played (or played around with) various instruments. I never had the time nor the obsessive dedication required to be a high-grade performer and I was perhaps more interested in improvising than in reading or performing scores.
During my lifetime I have listened to many thousands of hours of music, mainly classical. In my youth I voraciously absorbed works of mainstream composers but later gravitated toward the modern era, including jazz. In addition, I also enjoy early music, including Renaissance polyphony and Bach.
My own first efforts at sound art and composition began some time after my introduction to digital technology and online resources, which enabled me to intensify my listening and studies of modernist music. I began with various attempts at musique concrète. I acquired a digital sound recorder and wandered around capturing various sounds, anything from birdsong to building sites, together with recordings of domestic activities or apparatus, of small and found instruments, etc. At the same time I was investigating techniques and software with which to play around with the recordings I had made and published my first tracks online around 12 years ago. Since that time I have expanded my capabilities and resources. These days, although I still often use the techniques of musique concrète and recorded sounds, much of my work is purely electronic.
I have no room in my dwelling for banks of electronic apparatus or large musical instruments so that, together with small instruments, my equipment consists of laptops with Linux or Mac OS, with which I always use open-source software, for example: Pure Data (or “Purr Data” in its latest incarnation), Audacity, with hundreds of VSTs and plug-ins, PaulXStretch, and Samplebrain. I have an infinite array of sounds at my fingertips, both recorded and electronic. The challenge is to organize a few of them into interesting combinations and sequences.
It is not my intention to limit myself to one specific style or genre and I am ready to try anything that falls within the bounds of my competence. For this reason the Disquiet Junto perfectly chimes in with my activities. I also take part in the Naviar Haiku projects and, in addition to the music, have contributed the odd poem.
Some years ago it occurred to me that I could produce videos to post my music on sites such as YouTube and Vimeo, and I began to make fairly basic abstract animation to accompany the sounds. Gradually there came to be more of an equal balance between music and video. Now I often create the visual element before adding an appropriate soundtrack. I occasionally use music from Disquiet or Naviar projects in these videos, but more often than not the soundtracks are completely different and created specifically for use therein.
What is one good musical habit? Never discard or delete unpublished material, however unsatisfactory it seems. You never know if it might come in useful in a future project. There are no mistakes in composition: it is what you make it.
What are your online locations? I am not a huge fan of social media, although I do use Facebook from time to time. My music is available at soundcloud.com/ray_cobley and hearthis.at, videos at YouTube and Vimeo, music and videos at the Internet Archive (archive.org). All my work is published under a Creative Commons license and is free to download or remix.
What was a particularly meaningful Junto project? Rather than trawl through all my tracks and ponder over this for hours, I’ve chosen something fairly recent:
I have selected this because it presented a new challenge to me. I have no interest in dance music and, until I did some research, was not even sure about the exact characteristics of techno. In the end I was reasonably satisfied that I had created something slightly unusual or different in what I imagine can be a formulaic genre.
If I’m not mistaken, where you live is fairly isolated. Has the internet increasingly diminished any sense of isolation and provided a sense of community, in contrast with during your youth? With regard to quality of life, the Orkney Islands regularly feature among the best in the UK as a place to live. On the other hand, being distant from major conurbations and with a total population of around 22,000, the audience for my work is virtually non-existent here. A few years ago I arranged a presentation in a local gallery but hardly anyone turned up. Since this seemed to be a waste of my time and resources, I have not tried again. In other words, I depend on the internet to make connections. Fortunately, through the various online platforms and projects such as the Disquiet Junto, both as creator and listener, I certainly feel part of a vibrant international community.
As for my youth, life was so different in those days that I find it impossible to make comparisons.
Did you learn the lesson about not discarding anything the hard way? That is, do you have a memory of having thrown out something and later regretted the decision? Yes, that did occur a few times and occasionally I still delete tracks in error. By the same token, once something is published I try to make sure that I remove it from other files. I once found that I had uploaded the exact same track twice on SoundCloud under different titles!
The sounds in your beautiful recent YouTube piece, “Palette (Pure Data),” sound like field recordings to a degree — like fog horns, but transformed. When you are making sound from scratch in Pure/Purr Data, does your work with field recordings inform it? Upon consideration, I make little distinction between recorded and electronic sounds in the processes of composition. The origin of the material is less important than its destination. In other words, the main value is in the intrinsic characteristics of each sound or sequence and its potential for development or inclusion in a given work and, like many other artists, I often combine sounds from various sources in a single track. Looked at in another way, electronic sounds can be made to resemble recordings (as you suggest above) or vice versa.
Pure Data is a multi-faceted and highly adaptable tool and it is certainly possible to create sounds which simulate some that are heard in nature or field recordings. I have made use of this from time to time and, of course, on occasion such sounds can arise quite unintentionally. Any sound file can also be imported into Pure Data for use in conjunction with electronics but this facility is probably more useful in live performance.
In general terms, since I started with field recordings and musique concrète, there is no doubt that my experience with them has informed my later work in the sense that I have continued to use many of the structural and manipulative techniques and processes I learned at that time with all other sounds.
35: Percent faster that toddlers fell asleep in nurseries when classical music was played in a study
19: Percent faster for preschoolers
25,000,000: Fine, in $US, Amazon was hit with for the Alexa voice assistant reportedly violating child privacy
Sources: sleep (npr.org — thanks, Rich Pettus!), Alexa: (boston.com). And yes, “classical music” is a broad term. Presumably they weren’t playing Rachmaninoff.
