A Little More Human: Voiceover and AI

A Little More Human: Voiceover and AI

“The report of my death has been greatly exaggerated.” – Mark Twain

The same can be said about the demise of real human actors in voiceover. Despite lots of handwringing about voiceover artists losing careers to AI, there’s reason for us to be optimistic about the continued use of human voices, while acknowledging the reality of AI,  in the voiceover world.

First some reality. AI has been around for a while – hello, Siri, Alex et al – and is only getting better over time.

And it’s true: some clients will employ AI voice libraries – which have been recorded by real humans, by the way – for their projects. Some of them are very good. Apple, for instance, has some great sounding AI voices, at least for the 30 seconds I listened to their sample. Can these synthesized voices sustain a ten or twelve or fifteen-hour audio book? It’s to be seen if we can be engaged, informed and entertained by long-form AI narration, which is filled with pitch, pacing and emotional changes, not to mention the addition of different character’s, all created by the author with a different voice or point of view.

Anecdotally, several voice actors have confided they lost long-term phone IVR or eLearning projects to an AI voice library, but other actors report clients who eschew AI voices in favor of human voice acting.

Come what may, in the words of a fabulous speaker and writer, Catherine Johns, “I am going to double down on being human.”

The human voice has a voice print. Just like a fingerprint, our voice is unique to us. One synthetic voice may be one used by hundreds of other companies for scores of applications, which may get a bit monotonous. And who wants to sound like everyone else?

Language is quirky. It’s filled with abbreviations, acronyms, regional pronunciations. I’m from Massachusetts. We get a good laugh hearing out-of-staters pronounce names of places like Worcester and Cochituate.  Don’t get me started on what it takes to correctly pronounce medical and pharmaceutical words to make them sound natural. Lengthy editing and manipulation must go into making that audio sound like a real person who sounds like they are familiar with their surroundings, landmarks, and products. 

No matter how slick production values are, AI voices don’t sound human. They lack the pitch changes, pacing and anatomy that creates the ‘humanness’ of a human voice. My good friend, colleague, and voice actor advocate points to the ‘performance’ aspect of what we do. You can hear more of our conversation with Kirsten Rourke on AI, speaking and a wine recommendation! on my Embark podcast. 

Emotional connection and authenticity are words we often use when we talk about how we want to travel in the world. What can be more connected and authentic than the sounds of our own voices conveying emotion to another person?

It’s naïve and unrealistic to ignore the rise of AI in our world, deny its usefulness and our growing comfort in using it. Of course AI voices will take their place in the VoiceOver industry and will create a new business model,  new economy and new opportunities.

For now, I’ll keep it real and double down on being human. 

Please share your thoughts on AI along with how or if it benefits to life and work in the comments section. Or reach out at liz@lizsolar.com. Thanks.