Sounds Interesting
The latest news about sound, listening and speaking
Another bumper edition of my (vaguely) weekly newsletter.
My app is coming!
On September 1, I will launch my very own app. Entitled ‘Conscious Communication’, it will teach anyone the skills of conscious listening, powerful speaking and masterful presenting. I have put all my work, knowledge and passion into this amazing app, and I’m hugely excited… this feels like having another child! Watch this space, because next week I will be opening the waitlist, and there will be launch offers!
And now the other news…
Voice cloning is now common and indistinguishable from the real thing, as I found to my cost
AI voice synthesis has crossed a threshold in the past six months. Voices generated by leading systems are no longer recognisable as synthetic, and the latency is low enough for real-time conversation. Companies are moving fast to commercialise the technology for accessibility, customer service, and audiobook production, but the issues of consent and authentication are lagging far behind. Deepfakes of specific individuals’ voices (including mine) are already circulating, and the machinery for detecting them is losing ground to the means of creating them. My very disconcerting experience of this is that a clone of my voice is being illegally used to narrate Instagram reels on the dangers of jet skis.
This is not merely a technical issue; it is a question of identity and trust. A voice is intimate. We recognise ourselves and others through speech patterns built over a lifetime. When that voice can be copied without permission and deployed at scale, something fundamental is stolen. The regulatory frameworks being discussed so far are too reactive. We need to establish consent and authentication as foundational principles before this technology is further devolved and your voice is no longer your own.
Birdsong is good for you, proved yet again
North Carolina State University has added further evidence of the benefiots of nature sound. In a study of 106 people published in Environmental Conservation on 17 July, researchers sat participants either in a restored riparian forest or a neighbouring patch of grey space, played them either birdsong or traffic noise through headphones, and measured stress, mood, and how restored they felt. Every wellbeing measure moved with both place and sound. The forest with birdsong produced the lowest stress and the highest sense of restoration, while grey space with traffic was predictably the worst. Encouragingly, adding natural sound to a bleak setting softened its effect, which is a practical argument for designing birdsong back into our cities. In my book Sound Affects I describe how the sounds of the natural world, birdsong above all, signal safety to a nervous system shaped over millions of years to listen for exactly that.
Whales are shouting, and that should trouble us
Reporting this spring describes pilot whales calling more loudly to be heard over the din of shipping, a vivid illustration of how far our noise now reaches. The ocean has always been surprisingly full of sound, but now we are creating a deafening cacophony with our mechanical noise. Cetaceans in particular depend on sound to navigate, feed, find mates, and keep mothers and calves together, but propeller cavitation, sonar, and construction are increasingly drowning those conversations out. Blue whales can communicate with their subsonic calls over distances of at least 1,000 miles (1,600 km), but the noise of more than 60,000 ships moving at any one time is so intrusive that the whales’ range is now reduced to as little as one tenth of that, with devastating consequences for these magnificent animals.
Forcing any animal to shout is not trivial. It costs energy, and it frays the social bonds that sound often sustains. Cutting them off from one another by creating an auditory fog is even more serious. In Sound Affects, I make the case that noise is a pollutant like any other, one we tolerate largely because we cannot see it. When it comes to the ocean, this is even more true: we not only don’t see it but also don’t hear it – and meanwhile cetaceans and many other species suffer. Unlike physical pollution, we could reverse this instantly by stopping the noise. But where is the will to do this?
Spatial audio comes of age
For a decade, immersive sound has carried the stigma of being a gimmick, the aural equivalent of 3D television, which as we now know failed miserably. But that story no longer holds true. Recent industry coverage shows spatial audio moving from a novelty to an expected feature across headphones, cars, televisions, and home systems, with the complex problems of head tracking and low-latency rendering largely solved. The July issue of audioXpress gives itself over to immersive audio for augmented and extended reality, while companies such as Audioscenic claim to be building speaker-based 3D sound that follows your head with no headphones at all.
The interesting question is no longer whether the technology works but what we choose to do with it. Like any tech advance, spatial sound will initially spawn lurid and largely useless experiences, but I can vouch for the power and potential of being immersed in music. At TED 2026 I attended a workshop that had me sitting in the middle of an orchestra while they played. The emotional and physical effects were overwhelming, and I realised how much the paradigm of separation between musicians and audience was costing us. Music doesn’t have to come from in front of you, and nor do reproduced natural soundscapes. Stereo existed only because of the limitations of vinyl and now we are free of that limitation. Sound that genuinely places you somewhere, whether in a forest, in amongst a band or an orchestra, or in a fantastic soundscape for a game, is more powerful and immersive, and it is now within reach of everyone. Of course, it’s not obligatory: those who prefer two point sources can of course stay with that. But this is a whole new level of experience to explore, and personally, I welcome it.
Most leaders are poor listeners, and it’s costing a fortune
Perceptyx’s State of Employee Listening 2026 argues that in organisations, gathering feedback must be paired with action and learning in order to be effective, but the research finds that fewer firms are listening strategically or continuously, largely due to budget cuts for HR. My main proviso here is that what this report calls listening is almost entirely staff surveys, which are a very diluted version of the kind of person-to-person conscious listening I teach to leaders. Surveys are certainly necessary, but definitely not sufficient. Only organisations in what Perceptyx calls Stage 4 (Continuous Listening and Action) “have a robust and flexible listening program integrated deeply into their overall business strategy,” and sadly the proportion of the 750 organisations it surveys that are classed in this category has fallen from 36% in 2024 to 20% in 2026.
Meanwhile, Gallup puts the global cost of disengagement at 8.8 trillion dollars a year, and survey after survey finds that only a small minority of employees rate their leader as a good listener. In my book How To Be Heard, I argue that listening is a skill we are never taught and yet expect to master, and that conscious listening, choosing where to place your attention rather than letting it drift, is the most underrated leadership capability there is, as anyone on the receiving end of a distracted or over-dominant boss already knows.
Architects are at last starting to design with their ears
A shift is under way in how buildings are conceived. One commentary on 2026 trends in architecture suggests that acoustic behaviour is being written into designs from the first sketch, rather than bolted on afterwards to settle complaints. It sees this as part of a wider move towards sensory architecture, in which how a space sounds ranks alongside how it looks. Eco-friendly spray-on treatments, aesthetically flexible formats, intelligent systems that adjust reverberation in real time and 3D-printed diffusers custom-shaped for the way sound actually behaves, are turning acoustics from an afterthought into a design language. This matters in the places in which we work, learn, and heal, where sound quality bears directly on concentration and well-being. We experience buildings with our ears as much as our eyes, and it is heartening to see the profession beginning to listen. For the past three years I have given a CEU for the American Institute of Architects on this topic, with over 1,000 architects attending each year. You can access the most recent version for yourself here.
A comeback nobody predicted: reel-to-reel tape
Vinyl’s revival is old news. The surprise of 2026 is the return of reel-to-reel tape, which some audiophiles now hail as the most faithful format of all. What Hi-Fi has been chronicling a small but committed movement back to a technology most people last saw in the 1970s, drawn by the warmth and immediacy of a first-generation analogue master. There is a lesson in this beyond nostalgia. The relentless convenience of streaming has made music frictionless and, for many, sadly disposable. Threading a tape, dropping a needle, or simply sitting still to listen restores the intention, commitment and attentiveness that a streaming playlist strips away. The format matters less than the attention it demands. Whether it is tape, vinyl, or a well-chosen pair of headphones, the point is the same: most music rewards people who give it their full attention, rather than treating it as wallpaper. Perhaps this is another tiny ray of hope in the gloom of shortening attention spans and dopamine addiction among the young. Irving Stone once said of someone: “His mind was like a soup dish, wide and shallow.” We can but hope for deepening vessels in the future.
The new dawn of voice communication
Voice recognition in AI is moving fast. According to Assembly AI, the voice recognition market hit $18.39 billion in 2025 and is projected to reach $61.71 billion by 2031, a 22.38% CAGR. Latency for voice-to-voice interaction has dropped below 800 milliseconds, close enough to feel like conversation rather than command.
The frontier in voice technology has moved from understanding words to understanding acoustic space. Analysts describe 2026 as the year of spatial hearing AI, in which devices separate a target speaker from overlapping voices in a noisy, reverberant room, a tough trick for human ears as anyone suffering the cocktail effect will attest, and one that machines have until now found impossible. The implications are profound. A machine that can pick one voice out of a crowd is a boon for hearing aids and live captioning, but of course also raises privacy concerns, because something that helps us hear better can just as easily be used to eavesdrop. As always, technology is a two-edged sword.
The continuing curse of open plan
The evidence against the open-plan office keeps mounting, and researchers are getting better at measuring exactly why. Recent work in the Journal of the Acoustical Society of America builds a model of how people actually perceive open-plan sound, identifying pleasantness, eventfulness, and a sense of emptiness as the dimensions that matter, and confirming that noise is the enemy of enjoyment and concentration. Sound masking, better absorption, and simple zoning all help, but the issue is that we designed these spaces mainly for increased density (with the excuse of improved collaboration, now disproved by research), while imposing a huge acoustic tax on focused work. In my book Sound Business, I make the case that noise in the workplace is not a minor comfort issue but a direct drain on productivity and wellbeing, and one of the easiest wins available to any leader willing to take the sound of their building seriously. You can still listen to my BBC Radio 4 documentary The Curse of Open Plan here.
The biggest audio winner of all
A decade ago, the podcast was a niche medium for enthusiasts. New data from Edison Research shows that 45 per cent of Americans now listen to podcasts monthly, and the format has become the primary discovery mechanism for spoken-word content across the English-speaking world. So what changed? Partly it was logistics, as smartphones and wireless earbuds made the friction of consumption near zero, but there is something deeper at work: control. The podcast returns agency to the listener. You choose what to hear, when to hear it, and how much attention to give it. Unlike the broadcast model, which imposes its schedule on you, the podcast respects that you get busy and distracted, and offers an interaction that respects that reality; also, of course, you can listen while doing something else, be it driving, working out or having a bath. In a world of insistent notifications and interruptions, that flexible and respectful contract between speaker and listener feels almost revolutionary.
The sound of a good night’s sleep
Sleep science has moved beyond eyemasks and white noise. A new generation of sleep soundscapes, informed by studies of circadian rhythm, temporal patterns in nature, and the acoustic properties of deep sleep, is entering the mainstream. Companies like Endel are deploying AI to generate personalised soundscapes adapted to your location, time, and even biometric data, Of course, sound is not incidental to sleep; it is foundational. Our ancestors fell asleep to the soundscapes of their environment, such as distant water, rustling vegetation, or the breathing of others nearby, and that primal context is built into our nervous system. The modern bedroom with its A/C hums and air or traffic noise is denying millions of years of evolutionary tuning. Sound that restores acoustic alignment with our biology is not a luxury; it is a necessity.
Misophonia finally has a brain signature
Misophonia, the condition in which everyday sounds such as chewing or tapping trigger an involuntary surge of anger or panic, has spent years dismissed as mere irritability. A cluster of peer-reviewed work in the first half of 2026 has changed that. A Human Brain Mapping study of 939 adults identified a disorder-specific pattern of disrupted connectivity around the anterior insula, the brain’s hub for reading internal bodily states, and that pattern did not appear when the same data were sorted by anxiety, depression, or autism. Prevalence estimates now run between 5 and 20 per cent of the population. For the first time, randomised controlled trials in January 2026 showed that acceptance and commitment therapy and cognitive behavioural therapy both reduce symptoms. The reaction is not weakness of will; the salience network fires before the thinking brain gets a say. Recognition in the diagnostic manuals still lags well behind the science.