Dynamic range

Dynamic range is the distance in decibels between the quietest and loudest parts of a piece of audio. Wide range means whispers and shouts both land, which is thrilling in a cinema and miserable in a car. Narrow range means everything sits at a similar level, which is what podcasts and phone speakers want. Neither is correct; it depends entirely on where the thing gets heard.

The word gets used for two different things. Content dynamic range is the musical or dramatic variation you just described. System dynamic range is what the gear can carry: the span from the noise floor up to clipping, roughly 96 dB in 16-bit and far more in 24-bit. Recording too quiet wastes the second one; over-compressing destroys the first. Compression, limiting, and manual level automation are the tools for narrowing it deliberately.

The cautionary tale is the loudness war: decades of masters squashed flatter and flatter to sound louder on the radio, which stopped paying off once streaming platforms started normalizing everything to a LUFS target. Now a crushed master just gets turned down and arrives sounding lifeless next to a dynamic one. For spoken word, aim for around 6 to 10 LU of variation: even enough to follow in a noisy room, alive enough to still sound like a person.