It is well documented that babies are capable of screaming at the frequency most guaranteed to hurt our heads and ears in order to get our attention, furthermore they have sound level output equivalent to an angry F-16. This is a product of millions of years natural selection where those of us who were best at getting mummy's attention at crucial moments tended to live longer. All well and and good, but do we really need this level of danger alert in a modern two-bed apartment? Do we need the siren to last for hours at a time? Most parents (especially new parents) know the answer is a big fat 'NO'.
The BOSE corporation have recently pioneered 'noise cancelling headphones' which combine the signal fed to them with the antiphase of whatever sound is in the surrounding environment (presumably picked up by a microphone mounted in each earpiece). The resulting sound is played into the ears and the phase reversed ambient noise signal cancels out the ambient noise entering your ears leaving only the desired sound. This is an astonishingly difficult little audio trick to pull off, but they have apparently done it fairly effectively - I haven't tried the headphones, but read the reviews and you can see they do ok.
Now, no one wants to wear noise cancelling headphones all the time, besides which we usually listen to 'ambient' noise - other people, the phone ringing, the tv, just about everything really. So we have to control the child's noise levels at source. If we could reverse engineer the headphone technology so that it could hang around a pendant on the child's neck and transmit the anti-phase of a yell, we could silence the child. Of course given the difficulties of the technology we probably couldn't totally silence baby, but even knocking off 50dB would be a good start.
Some problems:
The pendant would have to be a point source of sound within a wavelength of the highest audible frequency the larynx produces (about 2cm for 10kHz) so would have to be on quite a tight necklace - it would have to be elastic for safety reasons.
Baby usually cries for a reason, and we can't remove that single method of communication from him/her. There should be an electronic volume control that would automatically reset to 'full on loud baby' when baby has quietened down and gone back to sleep. Baby can then wake up mummy and daddy at the next crisis point, and they can turn down baby's volume while heating bottle / cuddling / cleaning up vomit.
To counteract the volume of baby, we would need a seriously small, wide bandwidth and LOUD transducer. But hey, the larynx can do it, why can't we? Batteries would have to be included unless anyone at the Halfbakery has invented a way of powering electric devices off babies (anyone?).
Finally - massive feedback problems, this will be the hardest nut to crack. I haven't started on this yet, but it will be uppermost in my mind going to sleep tonight. Any thoughts will be posted tomorrow.
There will no doubt be many other problems which you can all inform me of - but isn't the prize worth the effort?-- wagster, Jul 04 2004 Etymotic Earbuds http://www.headphon...roductID=0020100000Expensive, but beat the crap out of noise-cancelling headphones, in my obsessively sound-aversive opinion. [jutta, Oct 04 2004, last modified Oct 05 2004] Review of noise-cancelling headphones http://www.thetrave...r.info/9nov2001.htmNoisebuster, BOSE, Plane Quiet [jutta, Oct 04 2004, last modified Oct 05 2004] Active Noise Control FAQ ftp://rtfm.mit.edu/...e-noise-control-faq [jutta, Oct 04 2004, last modified Oct 05 2004] The current agreement seems to be that you can't generally noise-cancel without access to the source or the path of the audio. As [ldischler] explained it: "Drop two pebbles in a pond, very close together. At a distance, you will have some regions of destructive interference, and some of constructive interference."
As you explain below, noise cancellation works well for low-frequency noise (like an airplane's engine hum), and not so well for high-frequency noises (like BWAAAA!!!!).
See the other ideas in this category for many similarly impossible applications.-- jutta, Jul 04 2004 True, but if the two pebbles are less than one wavelength apart, they behave as one pebble. Interestingly the noise-cancelling car stereo has been achieved (Mercedes I think), I'll look for a link. I don't think it worked perfectly, but it did work. Due to the above mentioned effect, low frequency sounds with long wavelengths were cancelled better than high-frequency sounds which were liable to interference patterns.-- wagster, Jul 04 2004 random, halfbakery