Speakers produce sound by a chance in voltage being applied to a audio driver (like a paper cone with a magnet to make it vibrate). Small changes in voltage makes a quiet sound, big changes in voltage makes a loud sound.
The audio driver software that converts audio data into an electrical signal for the driver necessarily knows those voltages, since that’s its whole purpose. Add in some information about the audio driver to approximate how efficient it is at converting electrical signal into sound and then adjust for decibels being a logarithmic scale (So a sound being twice as loud makes it 1 bel or 10 decibels louder) and voila, the driver software knows how loud it is.
The trick for wireless headphones is getting that information from the driver to the phone or computer, since I don’t believe loudness is part of the wireless audio standard. But that can be solved by asking the user to install a special app that lets the phone talk to headphones differently than using standard Bluetooth.
Some are more accurate than others.
For something like Apple’s Airpods on an iPhone they know the power the headphones can output and they can calculate that against the signal you’re sending them. Accurate enough for this kind of tracking.
When you’re mixing device brands, there may or may not be that knowledge shared ahead of time, in which case it would be a best guess effort using a likely common specification. Probably still fairly decent on the whole but with a few outliers that aren’t properly represented.
Latest Answers