Next gen Xbox console will reportedly see major voice control improvements

One of the many awesome features of Kinect is the ability to control the Xbox 360 console through voice interactions. The problem, as anyone who's tried it will know, is that it really isn't all that good. Far from being a natural way to interact with the console, the system requires users to voice specific terms or phrases that appear on screen and it can often be quicker to use a controller or remote to navigate to what you want, than to spend time shouting at the console. 

But all that looks set to change with the next-generation Xbox. The Verge has learned that Microsoft has big plans for voice interactions on the new console, which will see the system improved to support much more natural speech recognition. Instead of relying on clunky commands such as "Xbox, Music, Xbox Music", the user will be able to ask questions and give instructions in a more natural style, such as "what's new on Xbox Music?" or "show me my playlists". 

The enhancements go even deeper than this. The new console is said to support speech-to-text conversion, offering a significant boost to messaging interaction on the platform.

Even more exciting, it's believed that it will also allow users to wake the console using a voice command, something which isn't possible on the current-generation Kinect. This will likely rely on the new Xbox offering a low-power standby mode, to enable the console to respond to boot up on voice-request. 

In recent weeks, we've heard extensive rumours about the new console, including the suggestion that it will require a permanent internet connection, and even reports of its purported specs. There have also been persistent rumours of an 'Xbox Surface' companion tablet, which would make launch alongside the console.

The next-generation Xbox - codename 'Durango', though often colloquially referred to as 'Xbox 720' - will make its public debut later this year; full details will likely emerge at E3, although as we've previously reported, it's widely believed that Microsoft may hold a launch event for the new console before then. 

Source: The Verge | Kinect image via PCmag

Report a problem with article
Previous Story

Nokia blazes trail to T-Mobile with new 'Flame' 4G Windows Phone

Next Story

Dell can still call off its leveraged buyout plan by Nov. 5th

22 Comments

Commenting is disabled on this article.

If they can pull this off that will be big IMO. I think voice along with touch are going to be part of the future.

As for the other rumored always on connection, I doubt it. MS sells it's Xbox in markets where people don't have that option. They'd be effectively cutting out a chunk of their own market with that move. That's why I think it's wrong info and just another rumor that poped out of nowhere.

Its not a secret that Microsoft is working on a super advanced Voice Recognition Technology that will interact naturally with its users.

Its nice to have some short of confirmation that is implemented with the next XBOX but expect the same tech to appear in the near future in both Windows and Windows Phone devices.

http://www.youtube.com/watch?v=iB6pWs46GY8

PS. Note that the above video was published and showed my Microsoft well before Siri's introduction.

Microsoft always has the worst speech recognition for me.
At least Google Now is able to understand some of my words...
never tried Siri but it should be better than Microsoft one still.

Anyway all of them sucks if you are an asian.

deadonthefloor said,

I thought it was asians who sucked at english.
Well some are good. Just that we have different accent does not means we can't speak.

I don't know. I know some Asians and have tried it on my Windows Phone (not sure if it's the same technology) and it works great. Not sure what the issue with yours is.

Quikboy said,
I don't know. I know some Asians and have tried it on my Windows Phone (not sure if it's the same technology) and it works great. Not sure what the issue with yours is.
I never tried Winodws Phone's. However, the one in Windows 7 (Not sure if there is a different on Windows 8 or Windows Phone), is really terrible. Also, due to asian names, it is impossible to call or sms someone using voice control.

Oh. Sorry, I misunderstood. I can see what you mean about voice recognition have issues with non-Americanized names in general. Maybe you should give feedback to MS.

If MS is famous for getting everything right on Version 2.0 then they have had more than enough time to get this right. I will be out of my mind, if we will once again be beta testers on the next Xbox to get all this right for the next version.

As it stands, I love being able to tell Netflix to pause a movie, and rewind etc. but the entire process is indeed clunky even though it works. MS needs to nail this one the first time. Especially since this won't really be the first time, this should be the result of more than a decade of voice recognition tech that has indeed taken backwards steps as of late.

Even though Windows phone's recognition is outstanding, SIRI still has it beat with the setting of alarms and calendar appointments, etc. MS has had plenty of time to get this stuff right. Just nail it, enough of the good enough.

This will likely rely on the new Xbox offering a low-power standby mode, to enable the console to respond to boot up on voice-request

I don't really understand how they could make such a feature work. Surely it can't listen to and reliably process everything being said within its immediate surroundings at all times. If you have to press a button prior to a voice-request to turn the console on then I could just have the former do the job and the latter would be pointless.

It reliably listens to everything at all times when it is on, all they would have to do is power the Kinect's mics and have the Kinect be able to process certain commands without the console being on and send a power-on signal to the xbox. The command would be something like "Xbox: power up" Something plausible, easy to understand, but not a common utterance in normal everyday speech

Sraf said,
all they would have to do is power the Kinect's mics and have the Kinect be able to process certain commands without the console being on and send a power-on signal to the xbox

It would have to process everything being said in the background for it to know what to filter out and what to accept as the correct command. Seems impressive to me if they can do that on a "low power mode".

By the way, currently does it work well with background chatter/noise, do you have to really enunciate the command and does it require an Internet connection to process speech?

I don't have to enunciate more than normal unless I'm using the Bing function to look for a tricky word. Unfortunately, I can't remember what it was that caused issue last time, as that was a while ago. I can tell you that it is very good at hearing "Xbox" even if I try to muddle it up. My friends and I often call it "Sexbox" and it will respond to that

Sraf said,
It reliably listens to everything at all times when it is on, all they would have to do is power the Kinect's mics and have the Kinect be able to process certain commands without the console being on and send a power-on signal to the xbox. The command would be something like "Xbox: power up" Something plausible, easy to understand, but not a common utterance in normal everyday speech

Ok, you are NOT thinking this through...

Imagine a Kinect device that is 'self' powered and ONLY has to response to one concise command when the main Xbox is off. When it hears this ONE specific voice command and makes a match, it then signals the console to turn on.

This could have been implemented on the Xbox 360 if a 'basic' voice recognition processor had been put in the Kinect itself. (Which it is already known WILL BE present on the next Xbox.)

This is also FUD, this kind of stuff is just like the "Surface RT Trade-in" rumor, mis/bad-information that gets spread with no direct confirmation, and everyone falls for it, just because it's on the internet...

I do hope so, but the outrage (nowt worse than us outraged nerds!!) is useful as it will hopefully dissuade MS from attempting such a thing. If we all turned round after hearing the rumour and declared it to be the best thing ever you could bet that they would implement it straight away and cut Gamestop etc out of the market entirely.

TL;DR -Probably not true, still important to stop them getting silly ideas in their heads!!

Is it just me that thinks voice control progress went backwards this generation compared to the last generation? I remember playing Rainbow Six 3 on the Xbox with voice control, and it worked really well on there; it practically became second nature to me. The closest I can get to that on the 360 is with EndWar. Microsoft effectively gimped that area by encouraging developers to only allow in-game voice commands on Kinect, which from my personal experience with FIFA 13, is a disaster.