Gain staging
Setting the volume correctly at every step of the chain so nothing distorts and nothing gets buried in hiss.
See it
What it is
Audio passes through a chain of stages: mic, preamp, interface, track fader, plugins, bus, master. Each one can add or remove level. Gain staging means setting each stage so the signal arrives comfortably loud but never touches the ceiling, then handing it to the next stage in the same shape. Get it right and everything downstream behaves. Get it wrong and you are either fighting noise or fighting distortion.
Practical targets for spoken word: record peaks around -12 to -6 dBFS, with an average nearer -18 dBFS. That leaves headroom for a sudden laugh or a raised voice without clipping, and sits far enough above the noise floor that you never have to crank a hissy track later. The gain knob on the interface does the heavy lifting; the fader in the DAW is for balance, not rescue.
The misconception: louder recording equals better recording. In 24-bit there is so much resolution that a peak at -18 dBFS is still pristine, and you can raise it later with a negligible digital-resolution penalty, provided the source noise is controlled. The catch is that raising later also raises whatever the mic and preamp already put into the file, so a quiet room buys you more than a hot meter ever will. Chasing 0 dBFS at the input buys nothing and risks everything. Second gotcha: plugins have gain too. A compressor with makeup gain can quietly shove the next stage into the red even though your fader never moved.
Ask AI for it
Set up gain staging for a spoken-word recording: target input peaks of -12 to -6 dBFS and an average around -18 dBFS, keep at least 6 dB of headroom on every bus, use interface preamp gain rather than DAW faders for the initial level, and compensate any plugin makeup gain so each processing stage leaves the signal at the same level it came in.