The first is a spectral shaping module that allows you to graphically modify, in real-time, the resulting spectrum. In this way, it is possible to filter out single harmonics and partials, or to easily separate noisy and harmonic parts of the sound. The spectral shaping can be manual or based on an adaptive spectral envelope curve that fits the current spectral profile.

There is also a modulation section that applies an LFO to the spectral profile, creating regular patterns following sinusoidal or sawtooth-like variations. Finally, a pair of randomness dials allows you to control the degree of phase and position randomness of the algorithm.

Sloom includes five different timestretching algorithms: “wide” (producing a large stereo field), “narrow” (which preserves the original stereo contents), “rough” (producing a more aggressive sound containing pitch artifacts and beatings), “robotic” (producing a single-pitched synthetic sound), and “hybrid” (a mixture between the “robotic” and “wide” algorithms).

It can operate on both live audio input, based on recording an input buffer of up to 5 minutes, or on sound files in WAV, AIFF, FLAC, or MP3 format, without length limitation.

Sloom has been conceived as a sound design tool to explore radical timbre modifications, easily creating textures for ambient tracks, video soundtracks, events, or art installations that can potentially span hours, days, or even much longer. However, it should not be used for precise, transient-preserving stretching, which is outside its scope of application.