0
$\begingroup$

There are a new noise cancellation headphones that almost fully cancel human noise.

What kinds of algorithms could achieve that ?

EDIT:by human noise , i mean talking. In general, previous audio cancellation headphones didn't work that well for attenuating talking.

$\endgroup$
3
$\begingroup$

Since I cannot comment, here comes a short answer.

As far as I know, there are microphones outside the earphone. The microphones record the outside noise and generate sounds with 180 phase shift to cancel the noises. The noises are canceled actively.

$\endgroup$
  • $\begingroup$ The 180 phase shift is used generally in such headphones. But what makes those BOSE ones unique ? $\endgroup$ – blacken Jun 17 '14 at 10:54
  • $\begingroup$ @blacken: What makes you think they are unique? $\endgroup$ – endolith Jun 17 '14 at 15:14
  • $\begingroup$ The link is a review of them, and the reviewers think that, probably upon using quite a few headphones. $\endgroup$ – blacken Jun 17 '14 at 16:29
  • $\begingroup$ @blacken The theories are simple and basic. The implementations are quite different among different manufactures. The good ones require many experiences such as PCB design, power management and etc. $\endgroup$ – richieqianle Jun 18 '14 at 2:05
1
$\begingroup$

Noise cancelling headsets work in various different ways. The main technologies of choice are

  1. Passive sound attenuation
  2. Active cancellation using feedback loops
  3. Active cancellation using feed-forward noise prediction

The later two require microphones some place. The best sound reduction can be achieved by using all three technologies in a highly coordinated way. For example, the active technologies are very efficient at low frequencies but perform poorly at high frequencies. Passive attenuation (if it's well designed) is very effective at high frequencies.

$\endgroup$
  • $\begingroup$ Could you please explain the differences between feed-back and feed-forward approaches? $\endgroup$ – richieqianle Jun 18 '14 at 4:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.