Tag Archives: voice-recognition

Segway Robot: Proto Autobot

Posted on by

It’s no Optimus Prime or Hot Rod, but the Segway Robot does have a vehicle mode and a robot mode, and maker Ninebot hopes it will be useful whether you’re on or off the machine.

ninebot_segway_robot_1zoom in

The Segway Robot is a variant of Ninebot’s Mini Pro, which works much like the trendy self-balancing “hoverboard” transport devices. But its Intel Atom CPU, programmable robot platform, and Intel’s RealSense 3D camera system gives it a variety of additional capabilities. For starters, it has both voice and object recognition, and can be integrated to home automation systems. Ninebot also plans on making interchangeable accessories for the robot, such as a pair of arms.

ninebot_segway_robot_2zoom in

ninebot_segway_robot_3zoom in

ninebot_segway_robot_4zoom in

ninebot_segway_robot_5zoom in

ninebot_segway_robot_6zoom in

ninebot_segway_robot_7zoom in

Here’s a live demo of the robot from Intel’s CES presentation:

Ninebot will sell developer versions of the Segway Robot beginning on the third quarter of this year. The company will also release its Android-based SDK to help robotics enthusiasts unlock its potential. We’ll see if there’s more than meets the eye with this humble robot.

[via CNET]

CogniToys Children’s Toy Uses IBM Watson to Talk to Kids: Baby Deep Blue

Posted on by

A couple of years ago, I talked about Supertoy, which claimed to be an interactive talking kid’s toy. Sadly, it seems that the development of the toy has stalled, leaving many of its backers frustrated. This week a similar product called CogniToys popped up on Kickstarter, boasting the technology of no less than IBM’s Watson supercomputer.

cognitoys_1zoom in

CogniToys work just like today’s virtual personal assistants such as Siri or Google Now. You press a button, ask something and it talks back. The difference here is that Elemental Path, the makers of CogniToys, are using IBM’s proven supercomputer to analyze kid’s queries and come up with relevant answers. Another difference is that the toy sounds like Yoda.

CogniToys is based on FANG, a prototype that was made last year by Majestyk; some of Majestyk’s members would go on to create Elemental Path. FANG was one of the winners in IBM’s Watson Mobile Developer Challenge. Here’s Majestyk’s pitch for FANG:

A pledge of at least $99 (USD) on Kickstarter gets you a CogniToy as a reward, but the toy has already been fully funded so there’s no harm in waiting to see what happens. You never know, it might go the way of the Supertoy and become extinct.

[via Gigaom]

Samsung Smart TVs are Spying on You

Posted on by

So we all have to admit now that Samsung is being a bit creepy. They say that they have put in a voice recognition software so that they can hear people and claim that this will help them make a...

Less.Mail: Would You Trust AI to Answer Your Emails?

Posted on by

From time to time, I get some emails that I really, really care about. For example, people who want to donate to the ballet company that I direct or people who want to lend me a Bentley for a week. I can’t say I’d be willing to hand control of answering those over to AI. I want that money, and I want that Bentley. Of course, Less.Mail by Robin Labs isn’t designed to answer all of your emails for you, or even do anything totally autonomously, so it might in fact be a great idea.

less mail 620x469magnify

Robin Labs created Less.Mail to just compose responses to emails, and promises that the responses will work for 80% of your emails. It might be more like 40% of my weird-ass inbox, but for people with more standard occupations it might just work. The upcoming Android app is designed to automatically read emails to you and allows you to respond with brief voice commands. The AI then crafts automated responses based on your answers.

The real question is: would you be comfortable with sending a response generated by an algorithm, or would you end up rewriting all of them because it somehow feels disingenuous?

[via Tech Crunch]

OK, Google Everywhere: Big Google Brother Is Listening

Posted on by

It’s not a myth that everyone uses Google all of the time. If you’re constantly using Google, then you’ll like the fact that it might soon be possible to access the “OK, Google” voice assistant feature from anywhere on your Android phone.

ok google 1 620x364magnify

The OK, Google functionality is handy, but you currently have to be on your home screen or in the Google Search app for it to work. It allows you to perform searches, set reminders, set alarms, and more. New rumors indicate that it might be making its way into apps, providing app-specific voice commands. Your phone would be constantly listening to you for this activation phrase, and then could provide contextual commands inside individual apps. What’s not clear is if the already existing commands would also be available from within apps, or just app-specific commands.

[via Android Police via gigaOM]

Wall-E Toy Robot Mod Adds Voice Recognition & Proximity Sensors

Posted on by

Adriá García of DIYMakers augmented a Wall-E toy by making it work with voice commands and giving it the means to move on its own and avoid obstacles. It doesn’t compact garbage or collect curios, but at least it can dance.

wall e toy mod by diy makers 620x348magnify

Adriá used an Arduino Uno as Wall-E’s new brain. Two infrared proximity sensors help the toy detect obstacles, actuators move its arms and head and two continuous servo motors power its tracks. Adriá used the EasyVR module for voice recognition.

AUTO, navigate to DIYMakers for more details on Adriá’s mod.

[via BonjourLife]

Nuance Dragon Assistant: The Dispensation of Smaug

Posted on by

At the 2014 Consumer Electronics Show (CES), voice recognition specialist Nuance unveiled Dragon Assistant, a voice-activated personal assistant for Intel-powered computers and laptops running Windows 8. It’s not as smart – or sexy – as Samantha from Spike Jonze’s Her, but it’s a start.

nuance dragon assistant 620x378magnify

Dragon Assistant works a lot like Siri, Kinect and other voice-activated assistants. To start using it, you just have to say “Hey Dragon.” From that point you can then use everyday language to open applications, search online and of course dictate text. The program’s voice can be set to either “British Butler” or ”American Female.” Aww man, no German Shepherd?

See that? Now we’re all Dragonborn. Nuance says Dragon Assistant is already available on various Acer, Dell, HP and Lenovo laptops, ultrabooks, tablets and all-in-one computers, with ASUS and Toshiba computers to follow suit this year. Like other voice assistants, you’ll need an Internet connection to use all of Dragon Assistant’s features, although it can still execute certain commands even when you’re offline. Check out the Dragon Assistant support page for more on that.

[via Nuance via Gear Diary]

Supertoy Talking Teddy Bear: Thunder Buddies for Life!

Posted on by

Talking toys are nothing new, but most toys simply spout off pre-recorded sound bites. Not Supertoy Teddy. He’s a teddy bear that can actually hold a conversation, i.e. it will listen to what you say and respond accordingly. It’s like Siri, but in a stuffed toy.

supertoy talking teddy bear

Supertoy Robotics, the company behind Teddy, also made Jeannie, a voice-activated virtual assistant app for Android. In other words, these guys know a thing or two about natural language user interfaces.

I’m not sure if the toy in the video was actually working or if that exchange was staged. What I do know is that the Supertoy Teddy will work like Siri and Jeannie. And it sounds a whole lot like Ted. When you speak to it, it will connect to several servers that will decipher what you said and send back an appropriate reply. It will supposedly understand 30 languages out of the box. How will it do all of that? According to Supertoy Robotics’ response to backers, you’ll need to place a mobile device inside Teddy to make him work. The device will run Teddy’s app and will be the one to connect to the servers.

The video below, which shows Teddy’s innards, should give us a better idea of how the toy will perform in real life:

Hopefully Supertoy Robotics will give a no-frills demo of the final toy to clear up any confusion and show people what it can really do. But if you’re already sold, pledge about $64 (USD) on Kickstarter to get a Supertoy Teddy as a reward. I wonder if Teddy knows the Thunder Buddy Song.

Honda’s HEARBO Robot Listens Better Than Most of Us

Posted on by

While more and more devices are beginning to be able to understand what we’re saying, one problem that we face is that not everything that we communicate is in the form of a verbal cue. With that in mind, the roboticists at the Honda Research Institute have been working on a robot that can not only understand words, but other sounds.

hearbo listening robot

HEARBO (the HEAR-ing roBOt) uses a technology called Computational Auditory Scene Analysis to distinguish various sounds from its environment. This allows it to not just pick out verbal communication, but everything from the beep of an alarm clock to the mailman knocking on your front door. It can then use this information to make decisions about what to do.

The specific version of the technology being used for HEARBO is called HARK, an open source library that breaks down the understanding of sounds into three steps: localization, separation and recognition. This allows it to not only distinguish sounds in a noisy location, but exactly where they are coming from. In fact, the technology can nail down the direction that sound is coming from down to 1 degree of accuracy. Currently, the system can distinguish up to four different sounds going on at once, but in theory more microphones could allow it to handle even more sources.

Here’s a short demonstration video showing how the robot can understand what’s going on despite the world’s most annoying alarm clock going off the whole time:

And here’s an example of HEARBO easily distinguishing music from human commands, identifying song attributes, and even dancing a little bit:

So why develop such a robot? The technology could be used for things like robot butlers that could automatically answer the doorbell, or do things like turn off the TV if the kids have been using it for too long.

[via iEEE Spectrum]