On Mar 18, 2022 BMO Lab hosted a discussion entitled ‘AI As Foil: Hacking the Voice’, with Fides and London beatboxer Reeps One (Harry Yeff), moderated by Natalie Klym and hosted by the BMOLab at the University of Toronto. The following text is from Fides’ opening statement. View the full discussion here.
I have been singing professionally since 1981. I trained as an opera singer, went to the the Banff Centre mid 80s for two years, sang with orchestras in Europe thanks to the composer Luciano Berio liking the way I sang his Folk Songs as well as my abilities with difficult music like Schoenberg’s Pierrot Lunaire — a piece that upset early 20th century ideas of pitch and texture in Western Classical Music.
After that I returned to Canada in 1990 to create and produce interdisciplinary art as well as premiere new opera. Contemporary opera by and large features fresh approaches to orchestral timbre and melodic structure but the use of the voice is still traditional.
I have also taught singing for over 35 years. The artists who come to me have a keen need to express themselves but more often than not they’d like to do this without having their voice betray them.
But the voice is designed to betray and my job is to make this betrayal seem normal, advantageous even!
An interdisciplinary team of scientists under the purview of Dacher Keltner – a professor of psychology at UCLA, Berkeley – has studied “vocal bursts.” These are the incredibly swift non-verbal sounds we make as humans. They convey what we might call pure emotion. They happen in a split second and we decipher them in a split second – the thud of a disappointed sigh, the explosion of a laugh, the grab and punch of a scream.
We make conclusions about human sound almost faster than we can process it.
When I teach students to sing I am amazed by how quickly they want to sound good – to live up to what singing seems to mean aesthetically. In the first split second of a long tone – they alter themselves from burst like impulse to their perception of what I might want or what their singing practice demands – or try to match what the socially acceptable human voice has become.
We hear the intense need that humans have to fit in vocally through the everyday quality of our speaking voices — we are excellent censors. We like to keep a grip on things – tighten your anal sphincter and see how that tightens your throat. We also know how to keep a lid on things – remember the last time you you felt like crying or screaming and how in that split second you might have pulled down on your soft palate. Perhaps you get a lump in your throat when you are stressed? That feeling is called Globus and it is related to how fear can interfere with your swallow muscles. Our means of vocal censorship and our systems of survival are one and the same. Trying to be polite, intelligent, under control or dominant becomes etched into our breath and voice bodies.
To free my students from this tyranny I teach them to trust the relief of a sigh.
I also explain that the efficacy of a moan – whether as a result of pleasure or pain – soothes the moaner while conveying pertinent information.
Sighs, moans, grunts and sobs are part of our innate empathy processes – how we attune to one another.
Let’s take pitch out of the Western musical frames of measurable frequency and the letters we’ve attached to that. Instead of D, D#, E, I’d like to demonstrate a body frame for pitch and human sound.
Start on the D below middle C and go up by semitones 3 or four for each fleshy word! Discover these qualities are your voice transitions from one non-verbal zone to the next.
GRUNT, GROAN, MOAN, SIGH, SOB, LAUGH, YELL, SHRIEK, SCREAM, SQUEAL.
These body sounds are neither symbolic nor arbitrary – they link directly to sensation or feeling and how we manifest those sounds. That’s honest. But they also remind us vigorously of the cultural and familial constructs we call emotion. Those can be biased.
My exploration of the body/voice marriage began in the 80s through meeting and studying with Richard Armstrong of the Roy Hart Theatre. This voice work is often called “extended voice” but Roy Hart underlined that multi-phonics, chorded sounds and peeps were not an extension of the voice but a reclamation of what should be normal within human sound making.
These values were further incarnated for me through giving birth at home – I could be as real as I wanted. And thanks to my two daughters the ensuing years were full of their remarkably effective, bonding and economical sounds. Sounds both tender and fierce: utterly free while remaining connected.
Non-verbal sound became more and more the source of my creativity. The composers who found this wide vocal palate interesting were often the ones working either with interactive software or creating tape parts out of sounds that were made through computer processing such as granular synthesis. Zack Settel, Barry Truax, Maurizio Squillante, Wende Bartley and Louis Dufort to name a few.
Singing and improvising within non-traditional sound worlds encouraged me to sing and improvise using a range of undomesticated female sound. Computers gave me leeway around the elemental female voice.
In the early part of my career this acoustical environment encouraged me to step past the epic, beautiful and socially confined sound of traditional opera – where women tend not to fare well thanks to unrequited love, madness, or death. Take your pick.
I describe singing as being on the verge of laughing and of crying while capable of murder. Or for the healthy aggression required to say NO directly and with ease. Balancing the energies of mad, sad and glad creates an integrated sustainable technique. But first of all the singer needs to reweave “rest and digest” and “fight or flight” in their breathing, sounding body.
Because human animals make music from the same raw materials used for survival the unmediated voice maps our relationship to the impulses of the autonomic nervous system. We have learned to hack its star nerve – the vagus nerve – through meditation, deep breathing, any relaxation technique that changes heart rate and saliva flow. The vagus has branches that go to our digestive track, our heart, our vocal cords, our ears and our pharynx.
What is this mysterious pharynx?
Fake a yawn! Or maybe you are ready for a real one…
All of that space – the top of the throat, the back of the mouth, between the ears where the back of the nose opens to the top of the throat – all of that is your pharynx. When it is spacious with permission to belly laugh, sob or scream it lines up beautifully with the windpipe.
The spontaneous sounds that a yawn leaks are utterly uncensored. More like a bird that uses 100% of its air when making sound. We humans hedge our vocal bets through withholding breath.
As for artistic practice…
In recent years the unpredictability of the yawn’s physical and vocal gestures remind me of the fundamentals of good improvisation. And when I remember the freedom of a yawn while singing a composed song it seems like I am making it up on the spot.
The yawn speaks to speed – both fast and slow. And through its companions — sigh or moan — I rediscover the slow lag of an emotion filled vowel.
AH or OH or OOO are stinky with feeling. The word moan sounds suspiciously like mourn. The elongated mournful vowel of a song becomes the “blues” for both listener and moaner.
I am made of music. And that music is made of me. The range of the sound – the curves and spaces between pitches – those things become the melody. So feeling – whether pleasant or unpleasant, whether high or low intensity – is what the great composers – the composers of songs that become “standards” – understand in their flesh and bones. When I am not stingy with my air as I flow a sequence of notes you can feel it and stand under its meaning.
Part way through a workshop I was leading at Brock University an embodied feminist philosopher said, not disguising her exasperation, I can’t breathe and think at the same time!
A very recent research paper out of Japan states that humans do not sigh when they read from the screen, though they do when they read from the page. Screen time may be preventing our diaphragms from ebbing and flowing as easily as they do at night when our brains are powered down … lurking in dreamland rather than simmering in “flight or flight.”
Something that has been underlined for me while teaching voice on Zoom during Covid is that not all voices are equal in microphone mediated space – the male voice almost always comes across louder and takes over even when I have the male singer lower his mic volume. Why? Because the microphone was developed by men with the average male voice as the model. It likes and even amplifies lower frequencies.
The mic has the potential to mask or interfere with texture and timbre – with a full representation of the human body. So does over-production in a studio. The top end of the female voice is almost always EQ/d. Much like the upper sob in the male voice has been culturally squashed in the every day – the part we call fals-etto.
Who gets to determine what a human voice is – even before it is captured.
Chopin had a very clear compositional voice. At the last of these BMO lab AI events I heard AI generated Chopinesque piano music. I can believe it for 2 bars, then I can believe it again for another 2 or 4 bars, and then I realize that I feel like I am chasing my tail…that the internal conversation it is leading me through is stop and start. With a real Chopin Nocturne I experience the brilliance of his body and mind spending a generous handful of minutes feeling its way through a melodic argument about what it is to be human in the darkness of night.