Making Auto-Room Acoustics Software (Not spam)

  • Thread starter Thread starter Grilled_Cheese
  • Start date Start date
Grilled_Cheese

Grilled_Cheese

New member
I don't know where to put this...but it seems like most of the experts in acoustics post in this forum.

I'm working on a programming project to auto test a room's acoustics. Similar to RoomEQ but automated, and I need your help with a few questions.

Basically, instead of having to use a test-tone CD and testing each frequency individually and plotting its response (which can take a whole day to do), it does everything itself.

This is how it will work (tell me if something needs to be changed).

1. Set recording device to microphone
2. Set playback device to speakers/monitors

3. Start generating SINE waves from 1Hz - 20KHz (or beyond)

4. Program generates a wave at say 400Hz
5. Records that wave directly to disk (not through microphone)
6. Records another wave file from microphone and compares the two

This tells the program what the WAV "should be" and what the microphone picked up.

Uses frequency analysis to compare the responses from the 2 files. It then plots it to a graph, and moves on to the next frequency (401Hz).

As you can see, everything is automated, so you can just setup your speakers/microphone, hit record and wait until it's finished.

My questions are:

  • Is using sine waves effective for testing each frequency? Or does it have to be pink noise?

  • Does this seem like a more effecient way of testing room acoustics than using a test tone CD and doing everything manually?

  • Would there be a point in allowing people to test in 24-bit as well as 16-bit?

Thanks for any help you can give me. Once I finish with this project I plan on distributing it as freeware. :)
 
> 1. Set recording device to microphone
> 2. Set playback device to speakers/monitors

Okay, but you also need to let people select Line Input if they have a proper microphone and preamp.

> Start generating SINE waves from 1Hz - 20KHz (or beyond) <

I'd start at 20 Hz, maybe with an option to start at 10 Hz. Lower than that is pointless.

> 4. Program generates a wave at say 400Hz
> 5. Records that wave directly to disk (not through microphone)
> 6. Records another wave file from microphone and compares the two

That too is not useful because your program already knows (or should know) what the wave is supposed to be.

> Is using sine waves effective for testing each frequency? Or does it have to be pink noise? <

Either is a valid signal source, but sine waves are probably easier to analyze in your software.

> Does this seem like a more effecient way of testing room acoustics than using a test tone CD and doing everything manually? <

Yes, but a good analyzer program will also show decay times, not just raw frequency response. None the less, if your ambitions for the program are only as freeware, any feature you can offer is better than none. :D

> Would there be a point in allowing people to test in 24-bit as well as 16-bit? <

No.

> Once I finish with this project I plan on distributing it as freeware. :) <

Excellent!

--Ethan
 
The data will be meaningful only at the point the microphone is at. And you'll be amazed how much things can change just by moving the mic a couple of inches.
 
Thanks Ethan, really helpful. :)

I'd start at 20 Hz, maybe with an option to start at 10 Hz. Lower than that is pointless.
Click to expand...

Yes, there will be options to change the frequency range to test, and also intervals between frequencies.

apl said:
The data will be meaningful only at the point the microphone is at. And you'll be amazed how much things can change just by moving the mic a couple of inches.
Click to expand...

I know, but (some) people still like to test their mixing environment, for setting up monitors, changing listening positions, etc. And with this they wouldn't have to spend all day doing it. :)
 
Another question.

Ethan, you mentioned
a good analyzer program will also show decay times
Click to expand...
.

I'm not real clear on that...

I don't have much experience with room testing, much less, other room testing software (which is why I'm here), so could you explain that a little?

I don't see how a sine wave can "decay" naturally? Would there have to be a volume envelope or something? What would be an effective way to show decay times?

Also, on average, when testing a room using test tones, how long does each tone play for? Split second? Couple seconds? ...?

Thanks for the help so far, I may be back with a few more questions as I continue to develop this software. :)
 
Back when Mr Sabine was inventing the idea, he'd play a note on an organ and then use a stopwatch to time how long it was between the time he took his finger off the key and when he couldn't hear the note anymore.
 
> I don't have much experience with room testing, much less, other room testing software (which is why I'm here), so could you explain that a little? <

It's all in this article on my company's site:

www.realtraps.com/art_etf.htm

--Ethan
 
You must log in or register to reply here.
Back
Top