Microsoft touts breakthrough in making chatbots more conversational

Microsoft today announced (opens in new tab) that it has created what it believes is the "first technological breakthrough" toward making conversations with chatbots more like speaking to another person.

Microsoft says that it has figured out how to make chatbots talk and listen at the same time, allowing them to operate in "full duplex," to use telecommunications jargon. The company says this allows chatbots or assistants to have a flowing conversation with humans, much more akin to how people talk to one another. That stands in contrast to how digital assistants and bots currently work, where only one side can talk at any given time.

The technology is already up and running in Xiaolce, Microsoft's AI chatbot currently operating in China. Using "full duplex voice sense," as Microsoft calls it, Xiaolce can more quickly predict what the person it is speaking to will say. "That helps her make decisions about both how and when to respond to someone who is chatting with her, a skill set that is very natural to people but not yet common in chatbots," Microsoft says. Another bonus of the breakthrough is that people interacting with chatbots don't have to use a "wake word" every time they speak during a conversation.

The company says it is working to bring the same tech to Zo in the U.S., along with its other social chatbots.

This comes as Microsoft is increasingly focused on its AI and machine learning efforts, gradually spreading intelligent features across its lineup of products and services. Last week, Microsoft announced a major reorganization focused on AI and edge computing.

Dan Thorp-Lancaster is the former Editor-in-Chief of Windows Central. He began working with Windows Central, Android Central, and iMore as a news writer in 2014 and is obsessed with tech of all sorts. You can follow Dan on Twitter @DthorpL and Instagram @heyitsdtl

  • And yet I need to say, "Hey Cortana" every time I need to add one item to my shopping list.
  • Lol Microsoft research breakthroughs never translate to actual useful products. I've read through so many of them many years ago and yet have to see any of them come to an actual product in the way they promised. Microsoft is good at making half baked ****, getting lazy putting power behind it, and then abandoning it altogether.
  • Did you ever stop and think that it's not because MS is not producing anything out of these research directly and are maybe MAYBE licensing the **** out of it... MS is one of the biggest tech patent powerhouse in the world... Making money out of them is what they do.... Meaning almost any tech product you buy from MS or even apple android products and a **** ton of OEMs generates cash for them with very low risk to them...
  • I'm waiting for a breakthrough in comments here. So far, nothing.
  • The usual trolls have to trash MSFT first. As always. This has terrific potential. I look forward to chatting with a bot about the strategies and tactics being used in a football game I am watching. This should be good!
  • Research breakthroughs can be difficult to translate into real world useage. Hopefully that's the case here. I don't need someone else's phone or whatever recording and analyzing everything I say in an effort to make a grocery list for me. I'm good.
  • This sounds like a major breakthrough that could potentially bury all the existing digital assistant platforms.
  • Trust me, it's blabla, in 5 years Cortana will be as dumb as a cucumber, compared to now where it is as dumb as a paperclip. Oh wait, they both lack brains.
  • It seems they are initially targeting at the social chatbot. Cortana has to wait. But the "full-duplex voice sense" will probably spread to everywhere eventually.
  • Yes. It is a milestone so to speak in AI. Digital assistants are just digital assistants for now...AI and machine learning is on the forefront of that, but it all ties back to the same roots. Now that there is more RnD into IoT and I will assume a plethora of all kinds of other news we never even hear about it may not be long before some of this trickles down to Cortana.
  • I somehow read 'catbots' and now I'm disappointed. I think catbots are way more important.
  • Obviously more important, yet another example of Microsoft missing out on an opportunity to capture market share.
  • Longtime coming.
  • I only use chatbots in Skype and the novelty wears off after a few minutes, as for help desk digital assistants I generally get frustrated with them not being able to solve a problem with simple yes or no answers, that said, my last encounter actually had me believing I was communicating with a human until I realised the punctuation was too perfect!
  • so we're seeing progress?
  • These development will be much bigger in China since typing is a pain. Future of voice interactions with computer like we see on Sci fi such as Star Trek will likely develop first in Chinese. Google is banned in China. That leaves MS to compete with local tech.
  • Banned? Didn't they leave voluntarily?
  • I do wish MS would focus on practical applications of AI. For example I use the "Read Aloud" feature in Edge eBook reader all the time. Why? Because I'm kind of a miser and won't pay for Audible. If MS could make that one feature tell the difference between past tense and present tense, for example read "read" as "red" when the context suggests past tense, and "reed" when the context suggests present tense, then I will become an instant believer in AI. And may finally understand why they're investing billions in AI over say... acting like a SOFTWARE company and giving us some MS authored apps to fix the app gap in the bloody App Store.