While Amazon and Google vie for primacy in the voice-based input to the “smart home” and popular appliances (as noted by Amy Stapleton in this pre-CES post), Nuance Communications has taken a great leap forward by adding more artificial intelligence and voice-based speaker identification to its Automotive Assistant. The technology, which is already available in certain 2016 and 2017 models from BMW Group equipped with Dragon Drive, makes highly personalized, voice activated services available through the car’s infotainment console.

Nuance’s approach moves cars ahead of many home-based devices and many mobile apps in their ability to detect who is speaking and provide highly-personalized or individualized services. The example provided by Nuance, which will presumably be demonstrated at CES, showcases how a trigger word or inbound text message awakens an automated Automotive Assistant which can then understand the words in context and take the appropriate set of actions.

An inbound text during commute hours saying “Let’s meet at the Ferry Building in San Francisco,” may cause the Automotive Assistant provide traffic details and ask “Would you like me to provide directions to the Ferry Building?” In addition it may ask whether the driver wants to respond by dictating a text message. The assistant can also add the event to the driver’s calendar. Nuance calls this “AI-enabled text messaging”.

Two features of Automotive Assistant demonstrate that Nuance has gotten beyond the basics of Voice Input and Natural Language Understanding to tackle some tough issues that have major impact on creating a positive customer experience. First “barge in”, which enables the speaker to interrupt the assistant at any time to provide a command or dictate content. The second is the use of voice biometrics to distinguish whether the speaker is the driver, the passenger in the front seat or one of the passengers in the back seat.

Biometrics-based speaker identification inside an automobile enables an individual to gain access to personalized services through a shared resource. The driver can complete her text messaging while the passenger taps into a streamed version of his “Summer 1967 Roadtrip” playlist. This type of simple, strong authentication is prerequisite for secure, personalized access to the variety of digital services that are ready to be offered through the so-called “Internet of Things” taking shape in homes and shopping malls, as well as in cars.

Meanwhile, in another part of the CES venue, voice command of the “smart home” and all of its appliances and entertainment units on full display. Nuance has a few dogs in this hunt as well. They are centered on Nuance Mix, which was introduced last year to make it easy for software and app developers to add speech recognition, NLU and conversational interactions to any element of the Internet of Things. Its showcase implementation builds on a partnership with Greenwave Systems via its AXON Platform, and there are several other instances.

This year, many CES attendees are trying to sort out the hype surrounding the IoT. With stunning improvements in recognition accuracy and natural language understanding, the “Conversational IoT” is becoming a reality. The addition of AI and voice biometrics to Dragon Drive and Automotive Assistant demonstrate the potential for the automotive community to take the lead in offering highly-personalized and secure versions of the Conversational IoT.

Categories: Intelligent Assistants

Tags: Speaker recognition, Voice-Control, Intelligent Assistance, Nuance Mobile