Skip to main content

Microsoft has patented a method for 'silent' voice commands to address privacy, social awkwardness

One of the most fundamental problems with phone-based voice assistants is how awkward it can be to use them in public. As much as I use my Harman/Kardon Invoke at home to control smart lights, get calendar info, and create Cortana reminders, I pretty much never use her on my Android phone for one pretty simple reason, it's just a bit ... weird, to do so in public at least. Apparently, Microsoft agrees, as the company has patented a module that can detect "silent" voice commands.

As described by the company's patent filing, the "silent" input method can detect whispers and extrapolate voice commands from the airflow created while mouthing words. The module can be placed in a range of devices, including smart watches, phones, a smart "ring," regular headset microphones, and even a TV remote.

Although performance of voice input has been greatly improved, the voice input is still rarely used in public spaces, such as office or even homes. This is mainly because the voice leakage could disturb and even annoy surrounding people in quiet environment. On the other hand, there is still a risk of scattering private information to unintended audiences. These are not technical issues but social issues. Hence there is no easy fix even if voice recognition system performance is greatly improved.Implementations of the subject matter described herein provide a silent voice input solution without being noticed by surroundings. Compared with conventional voice input solutions which are based on normal speech or whispering that use egressive (breathing-out) airflow while speaking, the proposed "silent" voice input method is performed by using opposite (ingressive or breathing-in) airflow while speaking. By placing the apparatus (e.g. microphone) of the apparatus very close to the user's mouth with an small gap formed between the mouth and the apparatus, the proposed silent voice input solution can capture stable utterance signal with a very small voice leakage, and thereby allowing the user to use ultra-low volume speech input in public and mobile situations, without disturbing surrounding people. Besides of air flow direction (ingressive and egressive), all other utterance manners are same as our whispering, so that proposed method can be used without special practice.

As usual, note that patents don't necessarily translate into products, but there have been a few rumors floating around recently that Microsoft isn't done thinking about Cortana-focused hardware. We'll just have to wait and see.

Jez Corden is a Senior Editor for Windows Central, focusing primarily on all things Xbox and gaming. Jez is known for breaking exclusive news and analysis as relates to the Microsoft ecosystem while being powered by caffeine. Follow on Twitter @JezCorden and listen to his Xbox Two podcast, all about, you guessed it, Xbox!

16 Comments
  • This is just brilliant 👏
  • Awesome, now if they had a band or something to add this as a USP.. oh wait.. Joking aside, I hope this means Microsoft is getting back into the wearable market. Because it's a key element in the ultimate surface experience with home automation. I'm still looking into a few things... don't have much time these days...
  • Now if only they'd patent understanding my australian/swedish/god knows what else accent ...
  • I was singing a dramatic rendition of Sir Elton John's Your Song and laughed out loud in the middle of it.
  • I wish they'd actually work on natural voice commands instead of having to learn how to speak the correct way so the digital assistants have any idea what you're on about. I've had the unfortunate pleasure of having to setup and use Alexa over the holidays, and it's really bad; from having to still use an app to do even basic tasks, the awkward setup of multiple users, down to simply non-natural command structure you have to learn with 'skill' keywords or worse having to ask another 'skill' akin to an app, to do something for Alexa.
    It's just a very very weird setup and something that just doesn't flow, it feels awkward and not intuitive and unless you're already well versed and have remembered all the proper ways of asking things to get what you want, it can feel pretty useless and stupid... and this is why people feel don't like talking to their assistant in public, less the need for private information.
  • You honestly think they're not working on that? Maybe it's just not as easy to implement as you would like it to be or think it should be.
  • Skills are generally handled by third parties. They'll almost always be somewhat clunky due to not having the same AI behind their own systems as Alexa or Google. Personally, setting up something like IFTTT can make for some more fluid commands, but its still limited. In any case, as others have mentioned, this is a ridiculously massive obstacle to overcome and is a really big step in AI. Honestly, the fact that we can talk to a machine and even *possibly* get a relevant response back is downright amazing and we're living in a world where technological breakthroughs are happening at a rapid pace. I mean, you're complaining that your A.I. powered assistant can't understand you that well. That's an awesome problem to have considering what was possible 10, 15, or 20 years ago. And I think people don't use voice assistants in public because its generally more of a private matter. Most people consider it rude when people have loud phone conversations, why would we think any better of loud conversations with AI?
  • I recently switched from my Microsoft Band 2 to a FitBit Charge 3. I love the Charge 3, but I miss having Cortana capabilities and music control. I really hope Microsoft can re-enter the wearables market with this tech. I would use Cortana way more with this.
  • Oh, please. With so many loons walking around talking on their phones in the palm of their hands like they're serving dinner at a restaurant, I can't believe people care about this. And I'm with others here who wish they'd concentrate on reviving Cortana instead of shoving into the background or making it Alexa's b***h.
  • There are a significant number of people who can't believe that the Earth isn't flat. You are just as wrong as they are. Perhaps, next time you encounter one of these "loons", you should look around and see how many other people are about not behaving that way.
  • I'm with you there. I mean even I do it. I'll have conversation on my phone whilst shopping (not in an elevator, metro, etc.. closed compartments) when I'm out and about. If I needed to setup an appointment or reminder I'd just pick up my phone and state my command. Why would or should I feel embarrassed by setting up a reminder if anyone else was around? If someone asked I'd just say I'm setting up a reminder. What's to fret about. I don't get it. But I'll say, I'm glad to see Microsoft hasn't left room when it comes to innovating on mobile devices. Hopefully they'll enter the market once again with their own products soon.
  • Dude are you always this cranky?
  • What I >want< from Cortana, long before this feature:
    1. a setting to silence Cortana's responses to voice commands. (This shouldn't take a patent, Microsoft.)
    2. evidence that Cortana learns from its wrong answers and unacceptable responses.
  • Exactly - or even just your most recent commands listed instead of dumb example commands like "who dies in Star Wars" that Big can't even answer properly. And it's not just Cortana. Voice assistants have a long way to go.
  • Great idea. I'm definitely in the camp of feeling awkward speaking full volume into an assistant on a full bus, and in the winter I don't always want to pull out my phone and use the app instead, so a watch or headphones with this functionality would be great.
  • This is exactly the reason I stopped using cortana app on my android after the recent update that emphasizes on using voice and moved the text input behind another tap.