It might be possible to build an MCU-based solution that reads ambient noise levels (digitized), and vehicle speed (OBD/etc interfacing is probably non-trivial) and have it send vol-up or vol-down commands via the serial port. The one potential problem is that there's no easy way to contain "runaway" adjustments, although I suppose the MCU firmware could also sample the outgoing audio stream and see if the RMS signal level is "too high" (and force some limits that way). And this scheme of course assumes normalized volume levels for all tracks (good #@$!ing luck!), although even that might be do-able if the MCU keeps track of the average output level, and slowly increases volume for consistently "quiet" playback. But what happens when a loud track starts after a quiet one?

On another note, notwithstanding that player-supported noise cancelling would kick ass, isn't it actually something better suited to a separate device wedged between player output and amp (or pre-amp) input? A small box that samples the noise "out there", inverts, and mixes into the live stream (all analog) seems like a realistic possibility. Then again, I didn't go to school for EE...

Bear in mind, I'm still an utter novice at the moment, having owned my first player for just a week. But both hardware and software hackage are of keen interest to me.

Note that you're responding to a thread in which the most recent post is from 2003.

Quote:
It might be possible to build an MCU-based solution that reads ambient noise levels (digitized), and vehicle speed (OBD/etc interfacing is probably non-trivial) and have it send vol-up or vol-down commands via the serial port.

When you're at the point of making an external box... then.... why bother with serial port at all? Just have two pair of RCA ins and outs and attenuate those. Then you've built a box that will work with any stereo, not just the empeg. And it would work more reliably, and it would be simpler to build.

I've discussed this in other threads, possibly ones more recent than this 2003 thread. I think in one, I even drew a diagram of the box in question, asking if anyone knew how to build the necessary electronics.

Quote:
A small box that samples the noise "out there", inverts, and mixes into the live stream (all analog) seems like a realistic possibility. Then again, I didn't go to school for EE...

Yeah. Exactly.

Why hasn't anyone built one of these yet? Wouldn't it make a mint?

Quote:
The one potential problem is that there's no easy way to contain "runaway" adjustments, although I suppose the MCU firmware could also sample the outgoing audio stream and see if the RMS signal level is "too high"

Naw. You just make sure that the noise sampling microphone input is located outside the cabin of the car. Stick it in a wheel well or under the back bumper or something. That way it picks up road noise instead of music sound. Very simple.

You just make sure that the noise sampling microphone input is located outside the cabin of the car. Stick it in a wheel well or under the back bumper or something. That way it picks up road noise instead of music sound.

Ummm.... no.

It will be picking up the ambient noise in the immediate location of the microphone, which will bear little relationship to the noise inside the cabin of the car.

The noise inside the car comes primarily from wind passing over the various external protuberances of the car, from the tires contacting the road, and to a lesser extent from mechanical sounds made by the car itself. These noises enter the cabin of the car after being "processed" by the material of the car itself, that is, the sound from the tires is conducted through the wheels, suspension, and body of the car until it resonates inside the car. A similar process occurs with the wind noise. In both cases, the "quality" of the sound (frequency, intensity, etc.) outside the car is not the same as what you hear inside the car.

If all you want is volume adjustment... well, maybe. But it would probably be easier to tap into the car's OBD system to get the verhicle speed, or, for about $10 you can buy a magnet and a sensor switch (used on bicycle computers) and do your own independent measurement of wheel speed.

I don't think noise cancelling would even be possible without losing a lot of quality along the way. If the microphone were located outside the cabin, it would not be telling the proposed box the proper sounds to cancel, and if it were located inside the cabin, it would be "hearing" sound from the stereo and cancelling that as well as cancelling the noise. Even if you fed the signal from the stereo directly into the box and told it not to cancel any of that, some of the stereo frequencies would coincide with road noise frequencies and get cancelled anyway.

Noise cancelling works with headphones, where there is no "competition" between internal and extrernal sound. I don't see it working in a shared acoustic environment.

But then, my ignorance of such matters is only exceeded by my willingness to expostulate on them, so who knows?

tanstaafl.

I don't think you'd get the results you desire. To cancel the sound that you hear you'd have to grab the sound that is actually hitting your ear. If you playback the inverse of any other sound there is a chance ( I think a guarantee? ) that it will cancel some sounds but you will hear all of the other sounds that you aren't hearing that it is trying to cancel. I think that will just amplify the problem.

In other words : The recorded sound of the tires from the wheel well will differ from the sound that you hear in the driver's seat. You won't be canceling the sound you hear.

He's just talking about raising the volume, not cancelling noise.

Ahh. Word, then.