In today’s smart home, there are a variety of connected devices that we interact with on a daily basis, but these devices typically operate in an isolated environment. Despite these devices being “connected”, they lack a central access point. The user will often be required to use completely different methods to interact with and access these various devices throughout the home. This leads to a fragmented experience for the user, and overall adoption rates will be slow. With the abundance of smart home devices that are available, including thermostats, home lighting, security systems, music speakers and more, it’s becoming increasingly important to unify them in a thoughtful, organized way. As a matter of fact, Gartner’s projections show that there may well be 20 billion connected devices by 2020, reinforcing the need for a central access point to these devices. This can be achieved through the use of speech technologies including natural language understanding, automatic speech recognition and voice biometrics, all of which enable intuitive spoken interactions and likely would not exist without artificial intelligence as we know it today.
And with IBC 2016 kicking off, it’s a great time to consider the television — it is ubiquitous in the context of the living room, and acts as the centerpiece in many homes. Enabling the smart television to be the central access point to all of one’s connected devices not only simplifies the user experience, but could drive user adoption of these new technologies. In fact, in current DragonTV deployments we are seeing exceptional retention, with more than 75% of users continuing to utilize the service after six months of usage. Instead of adjusting to many different interaction paradigms, everything can be brought into one easy to access home hub.
Last year’s IBC saw voice-enabled TVs as a key trend. While this is still true, so much has happened in a year to bring us to where we are today. Even then, it was clear that “voice innovation… could help solve one of the industry’s most vexing problems,” in making content easier to find and enjoy. Trying to address the issue of sometimes-clunky visual menus, voice is proven to simplify things.
In such a short amount of time, we have also seen significant advancements in areas such as machine learning, deep learning and voice print identification. All of this helps pave the way to simpler spoken experiences, and features that may not have been feasible not long ago.
Thanks to these advances in speech technology brought about by artificial intelligence, it is possible to quickly introduce custom natural language interactions to today’s end-users via prebuilt models. This can be used, for example, to control the many services and devices available within the home, which is a growing need based on the explosive rate of expansion for the Internet of Things. With the flexibility to address specific areas of interest, otherwise known as domains (such as controlling the TV, playing back music, or querying the weather), it is entirely possible to offer users a cohesive and widely applicable set of tools that help fuel spoken interactions throughout the entire home. Speech and natural language can drive access to TV content, but more importantly, it can serve as the primary UI for the smart home.
To illustrate the power and significance of the TV as the central access point to the connected home, as well as what is possible with the latest advancements in speech technology, consider the following scenario:
A user performs a search for TV content, saying, “find me comedies with Will Ferrell.” Then, perhaps they hear a knock on the door so they say something like, “who is at the front door?” This could serve up a live video stream from an integrated home security system. At this point, they see that their friend has arrived and quickly ask the TV to “dim the lights” to control their smart lighting system, making for a more suitable environment at the start of the movie.
To take this a step further, voice biometrics offers a truly personalized smart home experience, despite being in an environment where many connected devices are shared. With voice biometrics, it’s possible to detect and identify the individual speaking, meaning that in our previous scenario, the guest could access separate device profiles or benefit from an entirely different set of suggestions. With a single user interface on the TV, it becomes possible to interact with many disparate devices in a personalized and refined way.
We at Nuance are focused on innovating the TV experience – with optimized technology leverages the latest advances in machine learning and deep learning to deliver highly accurate speech recognition that is adapted to the operator’s custom domains, fully and quickly customizable NLU, and easy to use voice biometrics. Nuance is pushing future advancements in AI for contextual and personalized conversational dialog which will deliver an even more intelligent and optimized user experience.
Again, general advances in AI and deep learning have made it possible to now offer speech technology that can help bring together the estimated billions of IoT devices expected to coexist in the coming years. Now is the time to unify the smart home, and we feel the television is the best way to do this. Visit us at IBC September 8 – 13 to see our smart home and speech technology solutions at work.