SUBSCRIBE TO OUR BLOG
To enhance in-car location and voice services, Nuance and HERE are working together to advance voice-activated point-of-interest (POI) experiences for the connected car.
Nuance is using HERE’s map and POI data to optimize on a global basis its Dragon Drive connected car voice technologies.
Arnd Weil, senior vice president and general manager of Nuance Automotive, tells us:
“There are two layers in a voice recognition solution: the layer that recognizes human voice and the actual words issuing driver commands, and the other layer that validates the command against what the system expects, which is the grammar, language or terminology.
“With so many street names, towns, countries, cities, places in the world – it’s incredibly important to ensure that voice recognition system can truly understand and recognize these words. As such, we need a strong dataset of map and POI data to train and advance our voice recognition. Otherwise the quality and accuracy of the content delivered to the driver will be severely impacted. By leveraging HERE data, we are able to further ensure that the driver receives the most accurate, relevant content available.”
Thanks to advancing technologies, consumers have come to expect unrivalled connectivity and ease-of-use from their connected devices. If the connected car is to become the essential commodity it is expected to be, these expectations must be met – then surpassed.
Nuance’s vision, as Arnd notes, is to deliver a “true automotive assistant, something that offers far more than route planning and guidance.”
He adds: “By leveraging a combination of Nuance’s voice and HERE’s map and POI data sets, we can offer a truly personalized, comprehensive driving experience, delivering everything from favoured routes and guidance, to the results of the driver’s favourite football team through a conversational voice interface.”
Nuance will be leveraging the data to further enhance the Nuance Automotive Assistant with the aim of providing a system that allows the user to issue simpler instructions, with a more effective response.
“If you wanted to visit the Hotel California, you could simply say ‘navigate to the Hotel California’ and the device will know that you want to access maps, rather than music by The Eagles,” says Arnd. “If there is confusion, the system will be smart enough to ask questions to resolve the issue.”
This takes an enormous amount of work at the back-end, though the availability and intelligent use of spatial and contextual information will make a real difference to the user experience.
For example, imagine that you’re planning a theatre visit in the evening, but the weather has taken a turn for the worse. Now imagine a system that, when ordered to direct you to said theatre, takes you to the closest, most sheltered car park so you stay dry. Smart, right? This is the reality that the relationship is working towards.
A system that intelligently learns what you need and how to offer it, without prompting, is the goal for this relationship, and would be a milestone for the connected car.
“By working with HERE to synchronize our databases, knowledge bases and the profiles that are used, we can offer a comprehensive experience that better caters to every driver and delivers a more connected car,” Arnd concludes.
We’re looking forward to a more connected future.