Understand what you change in the sound.
Each tool serves a different task. The manual covers choosing suitable input, understanding parameters and reviewing results.
Noise reductionReduce hiss, fans and persistent background noise using a learned noise profile or adaptive estimation.Speech cleanupSpeech-focused noise reduction with adjustable blend and multiple processing modes.De-clickTarget short clicks, digital spikes and pops.De-reverbReduce late room reflections in reverberant speech.Spectral attenuationAttenuate an unwanted sound in a selected time-frequency region.De-clipEstimate repairs for short clipped peaks in overloaded audio.Peak reductionTame protruding sample peaks while leaving below-threshold samples unchanged.EQShape low rumble, boxiness and distracting frequency ranges.GainRaise or lower selected audio uniformly in dB.Loudness OptimizeLift quiet speech with noise-floor protection and a peak ceiling.Extract speakerFollow a chosen voice using a clean reference passage from that speaker.Speech separationCreate two estimated speech sources as new mono clips.Spectral fusionCompare and spectrally combine treatment candidates made from the same input.
Preview is your shared listening check.
Preview helps tune a tool over up to 12 seconds. Loop repeats the passage; Input, Result and Residual stay aligned. Residual is a difference signal and may contain useful speech too.
Apply treats the actual selection. A good short Preview does not establish quality across the whole passage. Review the complete affected section after applying.