April 2016, NAB – Voice and language which is the most natural modalities we acquired since birth. No wonder that we are searching for communication with devices using our voice and language. Nuance Communications, has developed products that provide speech and imaging applications, is a leader in this area. If a customer uses voice commands to talk to automobile there is 99% chances that it is Nuance technology.

Kenneth Harper, VP, Devices & Ecosystem of Nuance talked to us about last advancements in transforming the TV experience with voice.  Nuance has been working for some time on specific solution called Dragon TV that is going to have huge impact on the TV experience.

The TV room is usually a complex multi-person environment where the commands are mixed with conversations making voice control is a big challenge. While you are watching TV, your kids are playing or your other devices are talking to you.  There are some advancements to solve that problem – said Harper. One of them is called signal enhancement.  Nuance is looking at audio that is being recorded, usually from two different microphones that, depending on set up, use an integrated technology and solution that we use that can sit in the remote control or set top box.  The technology trys to determine who is actually speaking in the living room. When we determine who that speaker is, we put what we called “a beamer” on that speaker. Then as a post processing task from all audio that was recorded, we enhanced that audio and suppress everything else. It called signal enhancement.

Accuracy matter in TV experience and Nuance making it work well.
Nuance_NAB2016_slide11

TV is not one piece but multiple pieces. Nuance covers the entire spectrum for TV by providing the enabling technology to manufacturers and then integrate it this with specific hardware devices or specific solution. Now when the second screens are considered as TV as well,  a companion app is in use in that case. Sometimes it is at the set up box, sometimes inside the remote control. The customers have their preferences and Nuance follows their needs for both solutions– said Harper.

There is a difference between using TV and training Dragon for PC use to helping write an document or general input to a computer– mentioned Harper referring to our journalism work. TV uses mostly short comments that are fairly predictable. For TV there are certain things people going to do.
Nuance_NAB2016_slide7

There are usually short commands, 5-6 unique words, we know the vernacular. Those are things that can be optimized – stated Harper. If a customer search for ‘movies with Bill Murray”. That’s how it all comes together.
Nuance_NAB2016_slide10

The future of the living room is set. TV becomes the central hub of the home and voice is becoming the primary interface.