# Approximating SPL from dBFS?

I am attempting to try to find a relative measure of SPL using a value of dBFS. Stats from the audio data are derived from SoX using mean/max amplitude values from -n stat

I know there is no direct relationship between the two, but is it possible to calculate a relative SPL using the following formula?

20 * log10(V1/V2)

Where V1 = amplitude of the audio file

And V2 = reference level of human hearing threshold (uPa?)

Example:

20 * log10(0.0211182/0.00002) = ~60.4 dBSPL

You see that tree over there that you're barking up? It's the wrong one. There's nothing up there. Seriously.

You know there's no direct relationship between dBFS and dBSPL, but then you go on to hypothesize that there is.

You were right the first time.

The easiest way to look at this is that you are trying to compare two things with completely different and unrelated references. One being "Full Scale" and the other being "Sound Pressure Level". What's missing from your equation is a way of relating the two references. It's a big hole.

The only possible relationship would be that if you were to be able to measure a dBSPL value at a particular point in space due to a sound source being in the same room.... if that sound source were to reduce in level by 6 dB, you could reasonably expect the dBSPL level to reduce by a similar amount.

But, knowing nothing about the sound source, the room, the measurement device, the speaker and the position in the room, you're on a hiding to nothing I'm afraid.

• Thank you for the reply, Mark. If we knew the specifications of the recording device/microphones, are there any ways of obtaining say, a relative measure? Feb 27, 2019 at 22:10
• When lining up a monitoring room for sound mixing, general practise is to line up the level of each monitor device individually to a pink noise supply at -20dBFS so that it generates between 78-80 dBSPL/C/Slow at the mix point. Once you have that setup done, then you should have some sort of relationship between the audio metering and the SPL experienced at the mix point. Does that help?
– Mark
Feb 28, 2019 at 8:05