Stories
Slash Boxes
Comments

SoylentNews is people

posted by Fnord666 on Friday April 03 2020, @08:24AM   Printer-friendly
from the fill-in-the-blanks dept.

Google's WaveNetEQ fills in speech gaps during Duo calls:

Google today detailed an AI system — WaveNetEQ — it recently deployed to Duo, its cross-platform voice and video chat app, that can realistically synthesize short snippets of speech to replace garbled audio caused by an unstable internet connection. It's fast enough to run on a smartphone while delivering state-of-the-art, natural-sounding audio quality, laying the groundwork for future chat apps optimized for bandwidth-constrained environments.

Here's how it sounds compared with Duo's old solution (the first is WaveNetEQ):

https://venturebeat.com/wp-content/uploads/2020/04/waveneteq_120_ms_2_63b829581a3291c144a030639139c199.wav
https://venturebeat.com/wp-content/uploads/2020/04/neteq_120_ms_2_8e86d7b2061dfb964b845ebefc1aebd9.wav

As Google explains, to ensure reliable real-time communication, it's necessary to deal with packets (i.e., formatted units of data) that are missing when the receiver needs them. (The company says that 99% of Duo calls need to deal with network issues, and that 10% of calls lose more than 8% of the total audio duration due to network issues.) If new audio isn't delivered continuously, audible glitches and gaps will occur, but repeating the same audio isn't ideal because it produces artifacts and reduces overall call quality.

Google's solution — WaveNetEQ — is what's called a packet loss containment module, which is responsible for creating data to fill in the gaps created by packet losses, excessive jitter, and other mishaps.

"I can['t] hear you now."


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This discussion has been archived.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, Insightful) by DannyB on Friday April 03 2020, @03:58PM (2 children)

    by DannyB (5839) Subscriber Badge on Friday April 03 2020, @03:58PM (#978758) Journal

    It's not autocorrect for voice, it's plausible deniability for voice.

    --
    People today are educated enough to repeat what they are taught but not to question what they are taught.
    Starting Score:    1  point
    Moderation   +1  
       Insightful=1, Total=1
    Extra 'Insightful' Modifier   0  
    Karma-Bonus Modifier   +1  

    Total Score:   3  
  • (Score: 3, Touché) by choose another one on Friday April 03 2020, @09:15PM (1 child)

    by choose another one (515) Subscriber Badge on Friday April 03 2020, @09:15PM (#978866)

    What exactly is the difference? I mean no one has ever blamed autocorrect for a text message screw up have they?

    • (Score: 2) by DannyB on Saturday April 04 2020, @06:07PM

      by DannyB (5839) Subscriber Badge on Saturday April 04 2020, @06:07PM (#979088) Journal

      It's not auto correct that would be blamed. it's the sender who would be blamed.

      The sender could claim that auto correct mis-stated what they said.

      Google's new "sound better" could modify your words to not be the words you actually said. But the audio sounds like you saying it. So you get blamed. You could claim it was google, but people might be less likely to believe that than they would believe it was an auto correct mistake in text.

      --
      People today are educated enough to repeat what they are taught but not to question what they are taught.