Talking to Your TV

Album Cover: The Future

"The maestro says it's Mozart, but it sounds like bubble gum."
Leonard Cohen / Waiting For The Miracle

Posted on December 28, 2012 1:02 AM in Computers
Warning: This blog entry was written two or more years ago. Therefore, it may contain broken links, out-dated or misleading content, or information that is just plain wrong. Please read on with caution.

I spent the first decade of my career working on speech recognition. I helped design and build speech applications that worked in cars and on PCs, PNDs and mobile phones. Over the course of that decade, I also helped design and build many prototypes that never saw the light of day outside a hotel suite in Vegas or some anticlimactic conference room. Of all the speech applications I had a hand in developing, none were as exciting or as promising as those involving TVs.

In an article at The Atlantic, Alexis C. Madrigal argues that television interfaces are getting steadily worse and wonders if TVs aren't the perfect place to realize the potential of speech recognition:

What I want is a simple TV controlled by my voice. Here's the only use case I care about: I want to be able to lie down on the couch, say, "put on HBO," and have the TV bend to my will.

Anyone who follows my blog doubtless knows of my disdain for Comcast. Though the reasons for that are myriad, one is that their TV guide interface is (and almost all TV/DVR manufacturers' interfaces are) of ridiculously poor design. Beyond that, the sheer amount of data that a TV guide conveys (or attempts to) makes finding a particular TV channel or show laborious.

Imagine speaking into a microphone in your remote control or just speaking in the general direction of your TV and saying things like:

  1. "Put on the Seahawks game."
  2. "Change to HBO."
  3. "I want to watch the highest rated movie playing right now."

Those tasks, particularly #3, can take several minutes with a traditional remote control and TV guide interface, simply due to the sheer volume of content that is available in a typical cable package these days. Now imagine you want to do one of those things on a friend's TV. With speech recognition, you accomplish the three tasks the exact same way regardless of where you are, what brand of TV you're using or what cable package you're dealing with. An interface like that, one that comes naturally to just about anyone who attempts to use it, is the kind of interface that really takes off.

Madrigal's article keeps the bar pretty low:

We don't do tons of stuff with our TVs. We want the channel to change. We want the volume to change. We want to record things. We want to play things. We want to turn the thing off. That's pretty much it.

...

The number of verbs the voice control system has to know is tiny and so is the number of nouns.

Though I think changing the volume with speech recognition is more painful than just hitting the volume down button, and though I know enough about speech recognition to realize it's about more than just verbs and nouns, I do agree with the overall premise that a simple, speech-enabled interface would make the TV user experience much more approachable. However, it's what I've seen behind closed doors that leaves me more optimistic about how speech recognition can (and will) breathe new life into the TV experience.

Unlike speech interfaces in cars, where emphasis is given to keeping both eyes on the road, the TV offers a visual and aural environment where speech makes a lot of sense. Some of the better visual TV interfaces, like TiVo (what happened to those guys?) and Windows Media Center, feel so much more powerful when responding to voice commands and instantly jumping from one screen to another as you say one channel name and then another. It's a flashy, instant type of feedback that you just can't compete with using the buttons or, gods forbid, the letter keypad on your remote control.

I'm very much looking forward to the day when someone steps out of the box and decides to make speech recognition a first class citizen in the television experience. I've seen glimpses of the future and I'm highly confident speech-enabled TVs will be a game-changer.

Comments

kptdhktf on May 15, 2017 at 5:22 AM:

Post Comments

If you feel like commenting on the above item, use the form below. Your email address will be used for personal contact reasons only, and will not be shown on this website.

Name:

Email Address:

Website:

Comments:

Check this box if you hate spam.