Stories
Slash Boxes
Comments

SoylentNews is people

posted by martyb on Friday August 26 2016, @12:11AM   Printer-friendly
from the no-sample-bias dept.

A Baidu voice recognition program has outclassed humans that were typing using smartphone on-screen keyboards:

Computers have already beaten us at chess, Jeopardy and Go, the ancient board game from Asia. And now, in the raging war with machines, human beings have lost yet another battle — over typing. Turns out voice recognition software has improved to the point where it is significantly faster and more accurate at producing text on a mobile device than we are at typing on its keyboard. That's according to a new study by Stanford University, the University of Washington and Baidu, the Chinese Internet giant. The study ran tests in English and Mandarin Chinese.

Baidu chief scientist Andrew Ng says this should not feel like defeat. "Humanity was never designed to communicate by using our fingers to poke at a tiny little keyboard on a mobile phone. Speech has always been a much more natural way for humans to communicate with each other," he says.

Researchers set up a competition, pitting a Baidu program called Deep Speech 2 against 32 humans, ages 19 to 32. The humans took turns saying and then typing short phrases into an iPhone — like "buckle up for safety" and "wear a crown with many jewels" and "this person is a disaster." They found the voice recognition software was three times faster.

Speech Is 3x Faster than Typing for English and Mandarin Text Entry on Mobile Devices (abstract) and full paper (pdf).


Original Submission

 
This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
  • (Score: 3, Informative) by Gravis on Friday August 26 2016, @01:16AM

    by Gravis (4596) on Friday August 26 2016, @01:16AM (#393272)

    This is only useful if the speech recognition has a large vocabulary because while you can type out "svelte," good luck ever getting speech recognition system to actually think you said it. Android thinks it's "smelt" every fucking time and it's using google to process this shit. Getting an offline speech recognition system that actually can handle your average 25k word vocabulary is just not happening, much less the whole 100k+ shebang. So don't throw out your zarf because if you move to speech-only input, you're going to get burned.

    Starting Score:    1  point
    Moderation   +1  
       Informative=1, Total=1
    Extra 'Informative' Modifier   0  
    Karma-Bonus Modifier   +1  

    Total Score:   3  
  • (Score: 2) by frojack on Friday August 26 2016, @03:07AM

    by frojack (1554) on Friday August 26 2016, @03:07AM (#393311) Journal

    Ok Google, how do you spell svelte.

    Works for me. The first time. Then most subsequent tries when to smelt or something close.

    Then I remembered languages that use sv at the beginning of words (Czech, principally) often sound them almost like two syllables, sss velt. Even though svelte is French or Italian in origin, it seems to work.

    After which it worked much better.

    But back on track, the story is just about the comparison to typing on a touch screen, and I have to agree with that.

      Voice reco is almost flawless for me. I feel utterly stupid reciting everything into a phone while everyone within ear shot is all up in my business.

    I imagine if you "wecomend a westuwant" you are going to feel left out of the talk to the device world.

    --
    No, you are mistaken. I've always had this sig.