Skip to Content

Nuance Thinks Voice Ads Could Be a Mobile Hit

Nuance hopes its voice-recognition tech can produce mobile ads that you actually want to have a conversation with.
April 1, 2013

In online advertising lingo, the acronym CPC refers to “cost per click”—the amount an advertiser pays whenever someone clicks on an ad. If voice-recognition technology company Nuance gets its way, though, it could soon have an additional meaning: “cost per conversation.”

illustration of man holding cellphone

Nuance is today announcing Voice Ads, a platform that will let companies create ads that people can talk to on smartphones and tablets. Mike McSherry, vice president of advertising and content at Nuance, says these could range from car ads that let you ask questions about the vehicle shown to ads for a sports network that allow you to get information about who won last night’s game or what time tonight’s game starts.

The company has lined up partnerships with several ad agencies including Digitas, OMD, and Leo Burnett, as well as with mobile ad distribution networks JumpTap, Millennial Media, and Ad Marvel.

Nuance, a company that was spun out of SRI International in the 1990s and now powers the voice-recognition capabilities included on billions of cellphones and in millions of cars, believes the time is right for mobile voice ads because many consumers have been primed by voice interactions on smartphones, such as with Apple’s digital assistant, Siri, and Google Voice Search on Android devices.

Meanwhile, the mobile ad space is growing rapidly as the popularity of smartphones and tablets skyrockets. Data from eMarketer indicates spending on mobile ads climbed to $8.4 billion last year from $4 billion a year earlier. This is expected to rise to $37 billion by 2016.

And while ad-targeting is improving most of the ads you see in apps and on mobile websites, they probably still mirror the banner ads, interstitial video ads, and search ads we see with desktop computer browsers. A number of companies are working on more creative ads for mobile devices (see “Google Searches Beyond AdWords”).  

On an iPhone, McSherry gave an example of what Nuance thinks would make a mobile ad more appealing: a voice-enabled ad for a fictional brand, Alpha Deodorant, featuring a talking magic 8 ball that has been pre-programmed with about 50 different spoken responses to account for various scenarios (including ones where it doesn’t really know what you’re talking about).

“What’s your question?” a stereotypical “bro” voice asks.

“Should I get a tattoo?” McSherry asks.

“How old are you?” the voice responds.

“I’m 44,” McSherry says.

“I would,” the 8-ball quips. “I’d get a Chinese character that you think means ‘serenity’ but really means ‘smells like butter.’ But just to hedge your bets, use Alpha!”

When the ad encounters a question it’s not familiar with—such as the question, “Should I have a baby?”—it does what McSherry calls “failing gracefully.”

“What’s the worst that could happen?” it asks.

McSherry can imagine brands like Disney using their character’s voices in ads, and suggests that a two-way dialogue with Mickey would lead to deeper engagement and brand affinity.

A startup called Volio, whose founder, Ronald Croen, also co-founded Nuance, has already developed voice ads that incorporate video. The company’s iOS app “Talk to Esq” lets you get cocktail, fashion, and hair advice from Esquire editors. It uses Nuance’s voice application programming interfaces, McSherry says.

For now, the dialogue in Voice Ads has to be recorded by voice actors, but McSherry thinks that eventually computerized voices—which now tend to sound stilted—will be good enough to re-create recognizable voices like that of sportscaster Bob Costas.

One vestige of existing online ads that will likely remain with Voice Ads is the idea of clicking. For fear of invading users’ privacy, the ads will not just pop open the microphone on your smartphone and start listening to you—rather, ads might have a little microphone icon that users click to start the conversation, McSherry says.

“There’s a mobile monetization issue. Everybody is looking for something to break it open,” McSherry says. “There’s been a targeting innovation, but we think we’re bringing a new interface innovation.”

But are existing mobile ads really performing poorly? Anindya Ghose, an associate professor at NYU’s Stern School of Business and co-director of Stern’s Center for Business Analytics, doesn’t think so. Ghose, who tracks the effectiveness of ads across smartphones, tablets, and PCs, says people tend to see ads on one device and may end up buying the product in question on another. He says this tends to happen with specific categories—shoes, clothing, and travel, for example. “It’s not that your ad dollars are being wasted. What it’s doing is nontrivially assisting sales of the same products through other channels,” he says.

Still, he does think we’ll see more attempts by advertisers to get users to engage with ads.

Nuance is banking on it.

“Most people don’t interact with ads for the fun of it, but that’s kind of where we want to get to with this,” McSherry says.

Jeffrey Bigham, an assistant professor of computer science at the University of Rochester and creator of a speech-recognition system that combines computers and crowdsourcing (see “Where Siri Has Trouble Hearing, a Crowd of Humans Could Help”), isn’t sure if Voice Ads can be made to work well enough that they won’t be frustrating to use. He notes that Siri, which initially appears to be able to do all sorts of things, is actually quite limited.

“You can’t just ask Siri anything. If you do, it sends you out to the Web,” he says.

However, since Voice Ads will be limited in scope by touting a specific product or brand, they may be able to be interesting and useful.

 “You’re not going to ask Ford about a good restaurant to eat at in San Francisco,” he says. “You’re going to ask it about cars. So that could make it work.”

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

Google’s Gemini is now in everything. Here’s how you can try it out.

Gmail, Docs, and more will now come with Gemini baked in. But Europeans will have to wait before they can download the app.

This baby with a head camera helped teach an AI how kids learn language

A neural network trained on the experiences of a single young child managed to learn one of the core components of language: how to match words to the objects they represent.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.