An alternative to touch screens? Voice control in the car is finally good: Ars Technica

Over the last decade, cars have become quite complicated machines, with often complex user interfaces. Mainly, the industry has added a touch to the near-ubiquitous infotainment display: it makes manufacturing simpler and cheaper and user interface design more flexible, even if there is much evidence that the touch screen interfaces increase driver distraction.

But as I’ve been discovering in several new cars recently, there may be a better way to tell our cars what to do, literally, by telling them what to do, out loud. After years of being, frankly, pretty rubbish, voice control in cars has finally gotten really good. At least on some brands, anyway. Imagine this: a car that understands your accent, lets you interrupt directions, and actually does what you ask instead of spitting out a “Sorry, Dave, I can’t do that.”

You don’t really have to guess if you’ve driven a recent BMW with iDrive 8 or a Mercedes-Benz with MBUX; admittedly, quite a small sample population. In these cars, some of which are also pretty decent EVs, you really can do without putting the touchscreen on for most functions while driving.

You can be general with your commands: if you tell the car “I’m cold”, it will raise the cabin temperature, for example. Or you can be specific: telling the car to “set the front temperature to 75 degrees” or “turn the seat heater to level 2” is, at least for me, much easier than remembering which segment of the screen tactile I am it’s supposed to sting.

The voice recognition is even good enough to understand me when I tell it to navigate to a specific address, to the point where I actually use the native navigation systems if I’m driving a modern BMW or Mercedes instead of relying on CarPlay like everyone else. . Having passengers in the car isn’t much of a problem either, which can’t be said about BMW’s gesture control when the front-seat passenger talks with their hands.

Some of that credit should probably go to Cerence, which supplies (among other things) the voice assistant to both BMW and Mercedes (as well as BYD, Renault, VinFast and others, Cerence told Ars) . Because so much of the software runs in the car, it has access to features that cars running Google’s Android Automotive operating system don’t have. Also, the previously announced Google Voice Assistant seems to have gotten worse at understanding speech over the last 12 months or so, for reasons I still don’t understand.

My enthusiasm for talking to cars seems to put me in the minority. Although a generation of nerds grew up with the adventures of KITT and Michael Knight, it seems no one wants to talk to their cars anymore. Part of this is a problem of exposure; as mentioned above, good voice control systems are not yet widely distributed.

But even among my colleagues who test the same cars for other outlets, I’m met mostly with skepticism when I praise good voice interfaces.

A 5,000 pound car is not the same as a smartphone

“I think part of it is just that there’s something inherently social about language. Over thousands of years it’s developed as an inherently social system. So I think there’s something about human beings that hesitates to speak with something that is not another sentient being,” said Betty Birner, professor of linguistics and cognitive science at Northern Illinois University.

“We’ll talk to our dogs, but maybe we don’t want to talk to our toaster. So I think that’s kind of it. That we use language to communicate and we have a sense of what communication means, and that means another mind. Right? My mind in communication with yours,” he told me.

“The other thing I mean, the obvious thing, is your car can kill you. Your toaster, I mean it could kill you, but you really have to work at it. With a car there’s a real danger, so you need to really trust the artificial intelligence there, and I think people don’t understand how far artificial intelligence and natural language processing has come, and they’re not going to trust it with their life. Which , you know, it’s understandable,” Birner said.

I read something that said voice interfaces were distracting

Also, if you look at the research, it tells you that, counterintuitively, voice interfaces can be distracting. A 2015 AAA-funded study found that it took drivers up to 27 seconds to regain full attention after giving a voice command to their phone or car, and another study by the same group that year tested various current infotainment systems and found its voice recognition abilities. be highly inappropriate.

“We think as designers that you should look straight ahead while providing voice input, but in reality, they found that people look where they think the microphone or the interface is, so they look away from the road when I’m speaking.” , said Dan McGehee, director of the National Advanced Driving Simulator at the University of Iowa.

Although technology has advanced significantly over the past decade, research on distracted driving has yet to catch up. “That was in previous generation systems, and it’s difficult. As researchers, what we should be doing is a longitudinal study to look at people’s attitudes and eye movements, eye tracking, while using these systems,” McGehee told Ars, adding. that funding bodies would want a pretty compelling reason to pay for such research.

Maybe it just needs time to filter through the public consciousness. “I think because the smartphone is really ubiquitous in our society and even around the world, people are slowly learning about these kinds of interfaces, whether they want to do voice-to-text — some people they use quite a bit, others don’t even. you know you can do it. So it just takes a while to adopt,” McGehee said.

Yelling probably won’t help

I don’t think we’ll get to the point of a voice interface for passenger cars, at least not in the foreseeable future. For some controls, especially those you need to use in an emergency, something physical is preferable.

“I think what was really interesting is that people feel like they have to shout, like if we’re talking to someone who doesn’t understand us, we tend to raise our voice, change our inflection. And of course that’s the worse. something you can do with a voice system,” McGehee said.

“In a previous life, I was a cabin designer for Boeing in their commercial division now over 30 years ago, and there they were trying the first attempts to use voice. And they discovered that, especially in situations where emergency, your voice changes and then those changes can affect how a computer interprets your speech and that was abandoned then just for that,” he told Ars.