One thing I've been wanting to explore for a long, long time was how to post-process recordings (of myself) to make it sound like it came from a radio. Recently, with the help of Google, I started doing it, and have been playing with some of my friend's recordings. At this point, I think I have found the essence of how to do it, and so would like to leave it here for future reference.
There are mainly just two parts to doing it: a compressor and a band-pass filter/equalizer. A compressor removes the changes in volume, so that everything sounds the same volume throughout the entire recording. A band-gap filter/equalizer attenuates certain frequencies to emulate the frequencies that vintage radios could or couldn't reproduce well.
In Audacity, the sound-editing application I'm using, both effects can be found under the "Effect" menu.
The compressor looks like this.
The main thing to note is the ratio - making the curve as flat as possible. The way things are set up above, for every 10dB change in the original volume, only change the volume by 1dB in the processed volume. This is done for all volumes between 0dB and -60dB (pretty much the entire range). This means that very quiet sounds sound almost as loud as very loud sounds. Finally, the checkbox will normalize the processed recording, by bringing all the volume up to 0dB, so that the final recording sounds appropriately loud. For more information on what each of the terms mean, I found this video to be very helpful at explaining things.
I apply two filters for the equalization. They look like this.
The first one is the preset "Bass Cut" filter. It is a high-pass filter, which diminishes the volume of frequencies that are less than 300Hz. This means that lower-frequency sounds (deeper voices) become quieter. For reference, typical adult male voices range from 85Hz to 180Hz and typical adult female voices range from 165Hz to 255Hz. In telephony, the usable range is from 300Hz to 3400Hz. More information can be found here.
This filter is a band-pass filter that diminishes both low and high frequencies, and only lets the middle frequencies (in the telephony range) go through without being attenuated. This is filter has the greatest effect on the final sound.
