Share

In 1968, Kubrick’s 2001: A Space Odyssey introduced us to HAL, an AI computer which interacts with – and eventually disobeys – its controller. 

Over 50 years later, ChatGPT, self-driving cars and the metaverse are just a few new planets in the solar system of artificial intelligence.

[AI is] no fad, it’s here to stay, and there has been much discussion on its place within adland.

37% of businesses and organisations employ AI and the total contribution of AI to the global economy is, according to a study by Gartner PwC Global, expected to hit $15.7 trillion by 2030. It’s no fad, it’s here to stay, and there has been much discussion on its place within adland (including this deepfake debate from the IPA). Brands are already using AI in their marketing strategies. Heinz, for example were able to prove that ‘Even AI knows ketchup is Heinz.’

But where do sound designers fit into the machine-learning jigsaw?

Above: Brands like Heinz are already using AI in their marketing messages. 

There are several arguments as to why AI will be unsuccessful in replacing human creativity and I’m inclined, fundamentally, to agree with each of them. A machine can only learn from the data it's being fed, it can’t take inspiration from conversations, books or films, and it is unable to make moral or ethical decisions. But, that said, the past few years have seen the emergence of several audio tools which are undeniably useful.

I know what you’re thinking, that this all sounds pretty ominous, but I don’t think that’s the case.

The rise of the humble podcast saw Adobe release Enhance Speech, a free (and brilliant) dialogue clean-up service. Just check out the before and after, you’ll be shocked. Building a makeshift voiceover booth out of a duvet is history – VO artists, rejoice! Another of their recent innovations is Firefly for video, a toolkit set to be implemented into Adobe applications which can storyboard and pre-vis directly from a script, apply a colour grade, compose music and add suggested SFX. 

Auto Design Desk, Autopod and Descript are just a few additional examples of plugins which can helps with previously laborious tasks, speeding up the track-laying process, turning type into talk (using your own voice) and cutting ambient noise automatically.

I know what you’re thinking, that this all sounds pretty ominous, but I don’t think that’s the case. In fact, I find it all impossibly exhilarating.

Above: Despite all its reported brilliance and industry-decimating impact, AI still has a long way to go.

You see, the reason people work with us sound designers, is because we’re translators. We support creatives to articulate their ideas through sound. The tools we use to do that have always evolved, but they remain secondary to that human touch. 

The role of the sound designer is going to shift, sure, but in a really positive way. We will become master manipulators, software puppeteers, delegating the boring parts of the role to the appropriate bit of software and slotting into a new position more akin to a creative director, orchestrating a suite of tools.

The role of the sound designer is going to shift, sure, but in a really positive way.

If you’ve got to this point and you’re still fearful of machine domination, perhaps the terrifying pizza ad generated entirely in AI [above], will pacify your rage against the machine. There’s a long way to go yet.

Craft and experience will always prevail, because of the nuance in human connection. Will AI negotiate a path between opposing notes from the director and the creatives? No. But to resist AI is to risk irrelevance. We must accept it, not reject it. Otherwise, we’re in for an odyssey of our own.

Share