Toshiba's auxiliary speaker cancels noise from main speaker
By Darren Murph 
posted March 19th 2009 9:22PM
posted March 19th 2009 9:22PM
posted March 19th 2009 9:22PM
A look back on popular stories from today in a specific year.
Now that we've thrown 'em off the trail, use the form below to get in touch with the people at Engadget. Please fill in all of the required fields because they're required.
to all those mac fans out there, http://img10.imageshack.us/img10/9823/1211544660807.png
Relevant... How?
Did you make this chart on Windows machine by any chance?
They need to add Pro Tools to that little sliver of Mac outside of the Windows blob.
I like how there's no overlap between OS X and Linux.
Things you can do:
Go away and stop adding irrelevant comments.
So let me get this straight: rather than make a good sounding speaker, they make speakers who's quality involves a bunch of distortion, then have another speaker to cancel out the distortion?
Haha, this Venn diagram was actually funny. I mean all OS's have their qualities and the first post of this article is no place for an image like that, but I enjoyed it for some reason.
now all we need is a speaker to cancel out the noise generated by the auxiliary speaker
exactly what i thought
Where will the madness end?!?!?!
The main speaker cancels out the auxiliary speaker.
Cone Of Silence anyone?
O
0
o
○
º
°
◦
•
·
^^^^ + vvvv = ----
so their making speakers that rid us of static and other unwanted noise... sounds good to me.
Not exactly,
Static comes from the source, so if the source has static, then you cant have the auxiliary speaker remove it.
This auxilary speaker removes the physical limitations of the speaker. ie. not being able to accelerate in time for a square wave, not being able to stop instantly.
With good source equipment, the only distortion occurs in the physical reproduction of the waveform.
I hope this clears things up a bit.
Basically they're making another useless technology for the uber-rich.
IMO.
This really isnt complicated, engadget. Did any of your writers pass high school physics?
This is the same exact principle that is behind basically every noise canceling headphone out there.
I used to read this site because I thought remarks like that were witty, but it is getting kind of painful as of late.
Came for this. Left satisfied.
99 percent of people have no idea how ACTIVE noise canceling headsets work. Apparently you don't even know the difference between noise canceling and active noise canceling so do us all a favor and chill. This technology is not used in "basically every noise canceling headphone out there" it's used in a very select few mainly Bose quietcomfort line and they're not cheap.
@p.flynn2
Totally agree with you. This is not that difficult of a concept. And rather start a flame war like Tyanney wants to do, I will try to enlighten readers with a little knowledge.
First, there are two kinds of "noise canceling" headphones, Active, and Passive. Passive ones use materials and good design to keep out noise, while Active ones actually employ a microphone to detect outside noise, and then using fancy circuitry, modify the signal coming into the headphones to compensate for the outside noise. I might add that passive headphones really aren't noise "canceling" at all, but rather noise dampening.
What Toshiba's technology aims to do is more accurately reproduce the signal being sent to the speaker. When a speaker makes noise, its vibrating according to the signal being sent to it in the form of electricity. But when things vibrate, they tend to keep vibrating after the signal has said to stop (inertia). To compensate for this additional unwanted vibrating, Toshiba is going to add a second speaker vibrating "opposite" the first.
Engadget, if you don't understand how this "canceling" works, look at the picture. I bet you can figure it out. If not, look up superposition of waves.
Whenever I see Darren Murph's name I immediately click the link to the original source, unless I'm in the mood to be baffled and confused.
This is the only reason I came into this thread. Done.
Would this is it was at all scientific and useful not need a $800+ microphone to actually see what the original speaker does wrong so it can fix it?
This whole thing sound to me like those 'magic' things the chinese like but the rest of us know is worthless nonsense.
Jwrigth, nicely written.
I know this is pedantic and I know why you put "noise cancellation" in quotes. The term gets applied when really "passive attenuation" should be used. You can't exactly "cancel" anything passively.
One man's noise is another man's music. Absolute linearity is not always preferred. How does it know it's not just picking up a complex harmonic anyway? This seems pointless and counter-productive.
Because, if I am correct, this is canceling out the line noise, which could be avoided altogether by a power conditioner. Why reinvent the wheel???
@brandon
That doesn't always work. Even if your speakers have power conditioning, that doesn't mean your sound card does. I think they're trying to mask systemic problems with a 20 cent gadget tacked on the end. The down side to this is that it's going to sound bad on a good setup, but that's not what it's meant for, right?
So you don't mind if your music is colored by the speaker's inability to accurately reproduce the input signal, and extraneous sounds coming from the speaker that are not in the input signal may not actually be considered noise by definition (it's not noise if you like how it sounds). Gotcha.
Power conditioning is not a solution, as a fluctuating power supply is not the only problem. Every component in the signal path introduces some kind of distortion, and there are many components in the signal path (the sound card, the speaker's electronics (ie. the amplifier), and the speakers themselves (ie. the moving bits, or the drivers)). Designers do their best to work around the limitations of components, but none are perfect even given perfect DC power.
Is this like noise cancellation on headphones, except on large speakers?
Kind of, but not really.
The technology is similar but the implementation and desired effect is very different.
if the phase is off by a tiny fraction (e.g. you dropped the speaker), will it double the noise?
It it was 180 degrees out of phase, it would double the noise and cancel the signal, but phasing is what the system is based on, and would be covered.
Interesting idea, probably near-impossible to pull it off. Speakers are not isotropic, and I suspect this will take some fine tuning to zero in on a specific "sweet spot" in the room to work. If they can get it to work well, it means better sound from cheaper speakers.
I like people like you; you always say what nobody else is thinking.
Fantastic. Seriously this means spending less money worrying about high quality interference protection on laptop sound cards. It's awesome! Cheaper solutions for everyone woooo
This is the technology that gives you sound reduction headphones Engadget where have you been ?
Car manufacturers have used it for years to cancel out road/engine noise this is not a new idea.
Toshiba want to use it to cancel sound that laptop speakers shouldn't be making, the buzz and the hiss.
Yet again no-one at engadget really knows anything about technology.
Wrong,
This is ridiculous, I am an audio engineer why would you add an extra device to deal with a problem caused by a crappy device. Simple solution provide quality components for the entire signal path and you don't need any noise canceling. Sony and Apple do it well, maybe Toshiba needs to out source its sound chip-set....
Wow, you must not be much of a sound engineer.
You of all people should know that no matter how good the speakers are, they are the weakest point in any system and are hindered by the laws of physics.
Odd, I never heard anyone saying what a wide soundstage or what astonishing detail and clarity Apple speakers produced.
Also, 'Simple solution provide quality components for the entire signal path and you don't need any noise canceling'...I am pretty sure that there aren't many speakers that are PERFECT no matter the quality of the components. $10,000 speakers with the system mentioned above would sound 'better'.
I am disinclined to believe cjharris13's claim that he is an audio engineer; if he were, he would know how ridiculous his remarks are. Heck, even I know how ridiculous his remarks are!
"This is ridiculous: I am an audio engineer."
Fixed!
If you were a real audio engineer, you would have realized that all balanced cables employ noise canceling before making that statement.
It's not that hard to understand. You can see how it works by using any sound file, converting it to a mono output and then using a new stereo file; place in each channel the mono file you just made and in one of the two invert the wave (upside down) with any editor, like goldwave for example. Then play the file you created from a set of two separate speakers, move them closer and see how sound "cancels" out. Same stuff noise canceling headphones do.
Either I'm missing something, or Toshiba is just trying to sell you a speaker that should really just be a small dongle attached to your main speaker wire. This could just as easily be done with a normal set of speakers by simply attaching some sort of noise detection device to one speaker, maybe inline between the source and the speaker itself. Since it obviously works through phase inversion of the noise played through the auxiliary speaker, it could work by detecting the noise before one of the normal speakers and simply adding the inverse phase noise into the main signal for one of the speakers. No extra speaker needed.
From the looks of it, they're just using differential signaling. This technology is already being used in USB (for data integrity) and noise cancellation. The science involved uses two channels:
Channel 1 (y) has the original signal/waveform (eg: y = x+1) *see 1st graph (Original Signal)
Channel 2 (-y) has the negated signal (eg: -y = -x-1)
Both of the signals are sent in phase (at the same time) with each other from the source (output of waveform generator). The transmission of signals from source to the input (speaker module) gathers noise. Noise can be attributed from your home power lines, running a blender, or turning on large machinery. Noise affects both channels uniformly, so:
Let's say noise = +2
Channel 1 with noise is now (y+noise = x+1+2 = x+3) *2nd graph
Channel 2 with noise is now (y+noise = -x-1+2 = -x+1) *3rd graph
Both channel 1 with noise and channel 2 with noise enter the speaker module. The speaker module decodes the original signal (y = x+1) by simple signal subtraction and amplitude reduction (by a factor of 2).
original signal = [(Channel 1 with noise) - (Channel 2 with noise)]/2
original signal = [(x+3) - (-x+1)] / 2
original signal = [x+3 + x -1] / 2
original signal = [2x+2] / 2
original signal = x+1 *4th graph
Now that the original signal is obtained, it is then sent to the speaker driver. The speaker driver powers the physical speaker to play the decoded waveform. That's it! It's quite simple actually!
I hope you enjoyed this,
Roman Sandoval
romansandoval.com
I think you may be going in the wrong direction.
See my above post.
the noise is a physical limitation of the speaker. And what you are talking about is re-active, this system would be calibrated and would be based on the input signal and knowing how the main speaker is going to react.