The startup trying to increase the reality of audio in public spaces

A startup called Spatial is unveiling its first product package today, focused on creating audio experiences that are immersive, interactive and automatically generated. The products themselves are a little complicated to explain, but the result is simple: ambient and interactive audio for public spaces that is easy to create and more dynamic than the usual tracks.

Although Spatial has a consumer offering, the most likely customers will be companies. Think of hotels that want a different audio experience in their lobby, theme parks that want to develop audio for their spaces faster, brand activations or AR experiences (National Geography is an investor). Think of how corny audio canned in the zoo is; Spatial wants to fix this.

In a demonstration a few weeks ago, I sat in a room and listened to the sounds of a forest around me, with birds singing in one place and then moving to another – until a dragon flew up and scared them for a while while .

Frankly, this is nothing special: the audio positioned in space is simple. What is complex is the mechanism that created all this audio. Spatial’s goal is to make custom sound design for a space easy and also to make that sound happen in a general way, rather than in a loop or track. There are three main parts to this.

First, there is Spatial Studio, a Mac application that is a kind of unholy fusion of Logic and Unreal Engine. It defines a 3D space where users can place audio objects – sounds that they themselves created or extracted from the Spatial library. Users can even stream live audio as an object – say, if they want to bring the sound of the ocean close to the lobby or just run a Sonos stream.

What is special about Spatial is that these audio objects have behaviors. A bird can move along a predetermined trail (with some randomness so it doesn’t get boring) or an ocean tide can appear at certain times of the day, for example. These audio objects can also react – to each other or to something that happens in real space.

The second part of the Spatial system is to transform all those dynamic audio objects into real sounds that you can hear in a space. To do this, he uses Mac minis (or, for corporate customers, Linux) to run a real-time audio engine called Spatial Reality. It will receive input from various sensors if necessary, or it will simply let the small world of audio take its course – and since things in that small world of audio behave differently, it will sound different all the time. Spatial has also created an iPhone application for more direct interaction.

You would think that the third part is the speakers, but in fact this is Spatial’s third trick: it can work with any speaker configuration. The Spatial engine acts as an abstraction layer that recognizes the position of the speakers in the room and automatically adjusts the sound to ensure the correct 3D positioning of the audio. Instead of a rigid set of location rules, Spatial can work with what you have.

Michael Plitkins, one of the co-founders, told me that he fundamentally believes that putting audio on a static track is the other way around. It is better, he says, to let computers find out in real time based on what they know about the speaker system. As the product is now, Spatial is not concerned with any real-time adjustment of the room’s audio. It will work with any speaker configuration, but users will need to program what they have in the Spatial Studio application.

In the beginning, Spatial’s main competition is some combination of Muzak for public spaces and any custom tools that Disney Imagineers use for audio in their theme parks. It may be attractive to some amateurs too – part of the inspiration for the company was Plitkins’ desire to create a soundscape in his own backyard. I had a demonstration of that space too, complete with cave sounds under the deck so authentic it was scary.

Whatever the customers, it probably won’t be an easy sale. (And launching a product aimed primarily at public spaces while a pandemic is still going on is another challenge.) Rain and birdsong created dynamically do not sound different from a static audio track if you listen for just a few minutes.

But during a long interview in a conference room where the team explained the intricacies of their product, they turned off the cabin in the forest – the audio with the theme that was playing smoothly all the time. The silence was strangely stressful, as conference rooms used to be.