In the context of computer music, it is important to have precise control of the spectral envelope of a sound e.g. for composers to realize their ideas or for the synthesis of the singing voice to localize formants precisely and thus to obtain good voice quality.
In the context of additive analysis and synthesis, where a sound is decomposed into sinusoids, rendering each minute detail of the sound accessible to manipulation, spectral envelopes can lead a way out of the dilemma of how to control the parameters of the hundreds of sinusoids over time in a sensible way.