Program for finding Equivalent Sound Level of an entire audio file

Gohanto

Gohanto

Veteran
So a non-film audio, technical audio question...

There's a project I'm working on now involving creating an auralization of a classroom to be built in the future (i.e. creating a sound file of what the room WILL sound like when it's created). The sound sources include a teacher's voice, and moped/weed wacker/airplane noise from outside. We know the teachers voice level and outside noise levels at 5 ft, 10 ft, ect...

Sound levels are general measured as "Equivalent Continuous Sound Pressure Level" or Leq, which is basically a logarithmic average of many sounds over a period of time (i.e. a quiet field might be 30 dBA for 59 minutes out of the hour, then fireworks going off at 70 dBA for 1 minute, probably an Leq around 62dBA-ish). We have meters that do this automatically that we leave outdoor for measurements, however now I'm trying to establish an Leq for digital audio files in terms of -dBFS. The reason for this is that if I can determine their (the voice and moped audio clips) average level, I can offset the audio files levels in the mix the correct amount.

In the end, our client can hear approximately what a teacher talking in the classroom would sound like while a moped is driving by, or an airplane is flying overhead, ect...

Yes, bizarre question. Any guesses would be appreciated.

Thanks,
 
Wow. This is one I haven't heard of.

Well, the first question is, what will the room be constructed from? Cinderblock walls? Wood? Drywall? What are the room dimensions? From here there are convolution reverb plug-ins that allow you to set up these parameters, and then you can record a voice, and throw that reverb on it. Then, on another audio track in your DAW, you can put in the other things, like moped by's, airplanes, weed whackers...but even then you need to know the outside construction of the building, where those objects are in relation to the building. Is there are a courtyard that the weed-whacker could be in? What about windows? What kind of moped? Engine size? How far away from the classroom are these audio sources? Unfortunately, I don't see a really accurate way to make it sound like how it will exactly sound like once the room is constructed in the real world. You can get a pretty close approximation, but, in my opinion, that's all it'll be.
 
Well, the auralization aspect is actually done (we use a program called EASE, run noise through an omnidirectional speaker at the wall with an open window behind it). We know the sound level of outside noise at the window, (67 dBA Leq) by having our meters propped out the windows for a week in the classrooms, and we know the sound level for a teacher at an average classroom distance (48 dBA).

So my question is, getting the equivalent sound levels of the digital audio files themselves, so we know how much to offset them. (yes probably gave way too much info)
 
Gohanto said:
Well, the auralization aspect is actually done (we use a program called EASE, run noise through an omnidirectional speaker at the wall with an open window behind it). We know the sound level of outside noise at the window, (67 dBA Leq) by having our meters propped out the windows for a week in the classrooms, and we know the sound level for a teacher at an average classroom distance (48 dBA).

So my question is, getting the equivalent sound levels of the digital audio files themselves, so we know how much to offset them. (yes probably gave way too much info)
Click to expand...

Wouldn't that be a function of the sensitivity of the mic? While manufacturers give some sort of spec, I think it would probably require that you actually calibrate the mic to the digitized output.
 
How calibrated is your output equipment? Because that will have a direct impact on the perceived volume of a digital file. I am at work right now, but when I get home this evening I see if I can break out the reference books for you.

But if memory serves, there is a formula to determine the dbfs offset for a given space. I could swear I had to do these calcs in school.
 
You can do Leq readings with the sound scope plugin in Protools. Also AudioLeak can do it.
 
You must log in or register to reply here.
Back
Top