Filtering a Sound

Filtering allows you to apply the effect of a frequency response (gain vs. frequency) to a sound. This procedure shows how to load a frequency response, visualize its shape then, how to edit graphically this response and apply it to a sound.

To Filter a Sound:

  1. Open one or more temporal signals.
  2. Click Tools > Filtering.
  3. In Signal, select by clicking the signal to be processed.
  4. In Filter, click Load then,

    • Choose Load from a profile to load a user-defined filter in *.spectre, *.txt or*.csv format

    • Choose Load a spectrum difference to load a spectrum difference calculated in Sound: Analysis and Specification

    • Choose Load a predefined profile to use one of the four predefined filters based on the human hearing perception: dBA, dBB, dBC or Inverse-dBA.

    • Choose Use current equalizer setup to use the frequency response of the current equalizer settings.

      The filter is loaded and its graph is visible.

  5. Use the design tools to modify the filter graph:

    • Line edition mode to edit the graph with the line mode

    • Free hand edition mode to edit the graph with the free hand tool

    • Shift curve vertically to reduce or emphasize all the frequencies in the same way at once (move the curve along the gain axis)

    • Shift curve horizontally to shift a peak

    • Stretch the curve horizontally to harmonically shift the curve along the frequency axis, keeping the rate between the frequencies and their multiple -the bass/low-pitch (towards the left on the graph) or the treble/high-pitch (towards the right on the graph)

    • Curve smoothing to smooth the curve (you can apply it several times in a row).

  6. If you want to apply the frequency response and open the filtered sound in a new temporal window, click Compute.
  7. If you want to apply the filter and save the filtered sound as a *.wav file, enable Save result in file then click Compute.

The sound is filtered by a filter having the frequency response that is displayed.