Held back by feasibility, technological limitations, and safety concerns, it might be a while before self-driving vehicles hit the road in numbers. But significant advances in speech recognition augur in a new generation of smart cars. And automakers around the world are listening.
Cars were never made for mobility alone. They represent utility, style and quite often, a big part of the owner's identity. We tend to make an emotional connection with the car, a romantic notion amplified in pop culture, where the driver talks to the car and, incredibly, the car listens. This connection isn't science fiction anymore. Cars are starting to understand us—and respond.
By 2022, says IHS Markit, over 83% of all vehicles sold will come equipped with some form of speech recognition technology. According to a Frost and Sullivan report, key developments in voice recognition include real-time translation, active noise cancellation, visual analytics to improve voice, virtual assistants enabled by artificial intelligence (AI), embedded natural language processing, deep speech, and voice biometrics. "Microsoft's Cortana and Amazon's Alexa are the latest offerings after efforts from Nuance in the personal assistant domain. Voice biometrics is expected to be a key element of personalization," the report says.
Leveraging speech recognition
The second most preferred method for a human-machine interface in vehicles after touch screen, speech recognition is now being leveraged for a broad spectrum of interactions, including:
Play music, turn on/off or search radio programs
Obtain news/weather reports (such as text to speech)
Set /change designation, like finding points of interest
Check/follow/share content on social media sites
General vehicle settings
Non-safety-related, such as turning on/off or adjusting lighting, temperature or open/close sunroof
AI-based personal assistant
Interact/command via speech technologies
Personalized setting/user account management based on a unique voice pattern.
Speech recognition is also seen as a natural progression in AI applications. While virtual assistants have been at home in our smartphones since Apple introduced Siri in 2011, they've now branched out in several new directions with powerful advances in tech by enterprises like Google, Amazon, IBM, and Microsoft.
Original equipment manufacturers (OEMs) and tier-1 suppliers are focused on providing intuitive user experiences as seen in the BMW HoloActive, which features a virtual touchscreen and a free-floating display. It's operated using air gestures, and tactile feedback is provided to the driver. A 3D digital instrument cluster with a camera monitors driver status and provides crisper graphics and enhanced voice recognition, adds Frost and Sullivan.
As the smart car becomes more connected, speech recognition technology will also help reduce accidents, as drivers will be able to effortlessly give voice instructions without manually manipulating onboard computer systems.
Around the world, speech recognition technology is being used by global car makers such as BMW, Mercedes Benz, Volvo, Nissan, Renault, GM Buick/Chevrolet, Ford, Honda, Mazda, Tesla, Geely, Great Wall and BYD.
Driving intelligent speech of the future
Last April, Flex inked a memorandum of understanding with iFLYTEK Co Ltd, a China-based AI solutions provider dedicated to the research of intelligent speech and language technologies, development of software and chip products, provision of speech information services and integration of E-government systems. Of these, intelligent speech technology is the core focus, which the company applies to multiple industries including automotive, medical, customer service, education, smart courts and smart cities.
Established in 1999 and currently employs some 6,000 people, iFLYTEK was listed on the Shenzhen Stock Exchange in 2008 and commands the largest market share among all businesses in the speech technology field in China.
The company entered the hardware and car devices sector three years ago and quickly developed a market for noise cancellation modules, with total shipments exceeding 1 million as of June 2017. The hardware team is now working on second-generation head units equipped with speech recognition and noise cancellation technology.
“One product Flex is manufacturing for the cars of the future is a noise cancellation module in collaboration with iFLYTEK," said Chris Obey, Flex President, Automotive. "The product is being manufactured for multiple OEMs and tier-1 suppliers, and promises to reduce noise inside cars, which is essential for voice recognition accuracy," said Obey.
He added, "A second product is a speech recognition-based head unit which we’re currently manufacturing for Chery. We’ve received very positive feedback that this product has helped boost the sales of cars it’s on."
While debates on utility, feasibility, and safety rage around self-driving vehicles, which depend on advances in AI, there’s nothing but momentum for AI applications when it comes to natural language processing in mobility solutions.
Speech recognition is poised to transform the landscape of the automotive industry, with far-reaching implications in every aspect of the automotive experience, much as 5G technology will soon be deployed to enable vehicles to communicate with each other, the surrounding environment and passengers.
With such interesting prospects ahead of us, we're eager to accelerate into the future.