Imagine you get into your car and automatically, your seat and mirror positions are adjusted for you, your favorite radio station starts playing, email and Facebook accounts are opened, your navigation favorites and past destinations are loaded, etc. This kind of personalization has been a chief goal of automotive UI design for a long time. I remember talking to a friend at a carmaker known for its innovative concepts. In the late 90s/early 2000s, the automaker’s cars identified the driver via an ID chip in the car key. It was effortless: you inserted your key into the ignition lock, were recognized, and the car adopted your settings. This worked well except for the cases when people accidentally picked up the wrong key. Say somebody took his wife’s key and she happened to be shorter, the helpful UI moved the seat to a position that was just perfect – for his wife. He would find himself hugging the steering wheel. The car key identification was quickly abandoned.
Enter speaker recognition a.k.a. voice biometrics. With no additional ID tokens like car keys that may be mixed up, the driver or passengers can be identified en passant while interacting with the car’s speech dialogue. It’s effortless and works with the hardware that is already available in the car.
Our voice biometrics technology can be used as standalone identification technique or combined with a wake-up word and speech recognition/natural language directive into a one-shot command. Take a phrase like “Hello Dragon, show me my email”. In one simple utterance, the user
- activates the dialogue through the wake-up phrase “Hello Dragon” at the beginning of the command,
- is recognized per voice biometrics as the legitimate driver of the car (and email account owner), and
- instructs the speech dialogue to present her email to her
It’s like having your own personal butler in the car with you that reacts only to your commands. Hands-free, eyes-free task completion with low cognitive load: a safe and natural automotive UI.
Voice biometrics can be used to develop an automotive assistant that serves multiple people, each in their own, personal way. When I enter my car and speak a voice command, “play my music,” I want it to choose a different playlist than when my wife speaks the same request, and my kids certainly have different thoughts on what music is cool than I do. With our software, such an assistant may be developed to only assist the driver, so, say, when you share the car in a family, every driver gets a personalized UI. But a car-maker can just as well develop personalized incarnations of the assistant for all people travelling in a car at the same time. So while a family is driving, dad may request “open my navigation destinations” and he gets his destination list, two minutes later, mom says from the passenger seat, “Hello Dragon, post to Facebook: ‘driving around Lake Lugano, love the place’” and the assistant posts to her Facebook page, and when the kids from the back seat scream: “We want to see movies,” they get their favorite films, because the assistant recognizes the speakers and knows each user’s preferences.
As systems become smarter, we expect them to be more helpful, understand what we want, and cater to our specific needs and desires. Voice biometrics helps to put the personal in the personal assistant.
Beyond personalization, the second big benefit of voice biometrics is security. The connected car is becoming more and more just another connected device providing access to personal content. When entering a connected car, you don’t only get control over the vehicle itself, but also to the data and services to which the car is a gateway. I’m happy to lend my car to my neighbor, but I certainly don’t want him to read my e-mails via my car’s infotainment system. With voice biometrics, I can be sure only I get access to my services if that’s how I want to control it. The same Nuance voice biometrics software that we use to protect your sensitive info in cars is deployed for bullet-proof banking security solutions.
Going forward, what if I can also pay at a parking lot or toll booth with my voice because my car verifies my identity per voice biometrics and provides a token of my identity to the payment system, with my voice as my password rather than having to mess with signatures or typed PINs and passwords? Maybe it’ll have keyless entry and I (and only I) can open and start my car per voice?
I’d buy a car like that. And chances are, once my car has gotten to know me and created a personal and efficient UI for me during its lifetime, I’d stick to that carmaker and get a car from the same brand again if it allows me to carry my settings to my next car, so right from the start, the new one recognizes me and provides me with something I want from my car: my own, familiar experience.