Once again, the use of the lossless mode is recommended for the effects below.ДЄs explained above, you can separate a sound's envelope from the rest of the sound, work on that envelope separately and then re-apply the modified envelope the your flat-envelope sound.

Also prior to loading the sound it is recommended that you either select Constant Time Resolution Analysis in the Options menu, or set min_bpo to 0 in the config.txt file, to ensure that the envelope matches more closely to the real envelope of the sound.Д¬ompression, expansion, limiting, noise gating The only changes in the techniques consist in using the vertical operations equivalent to their horizontal counterparts.ДЇor best results the use of the lossless mode is recommended. The vertical equivalent of the three equalisation techniques described above can be used to apply an envelope to a sound, correct the envelope of a sound or directly edit a sound's envelope. Knowing this you can transform Layer 1 to for example cut peaks or remove noises (for example by moving (using the Move tool) the layer to the left so that the background noise is cut off) and when you're done with editing this layer you can average it and make sure its blending mode is set to multiply to hear the sound with its new modified frequency profile.

At this point we've effectively separated one characteristic of the sound (its overall frequency profile) from the rest of the sound, and we could combine them back together again by averaging horizontally Layer 1 and multiplying it with Layer 0. This has for effect of making the frequency profile of that layer flat, in other words every band of frequency has the same intensity. Then, go back to the layer (Layer 0) that represents the original sound, and equalise it (menu Operations > Horizontal Normalisation). Starting from a freshly loaded sound, duplicate the sound and turn the duplicate layer (Layer 1) into horizontal curves as done previously. This is quite convenient for removing undesirable peaks in frequency. You can equalise the sound directly by editing its frequency profile curve. There is yet another equalisation technique, this time much more direct. Note that you may want to do another equalisation on top of it to give your sound the frequency profile you want. This causes the bands of frequency that were identified as weaker during the drawing of the curve to be boosted and the stronger bands to be dimmed to both about the same level. Once you're done, fill it as done previously and average it, but this time either set the blending mode to divide (menu Layers > Division) or keep it to multiply and invert (menu Operations > 1 / x, not to be confused with the other command named invert that instead does 1 - x) the layer. To do so simply draw a curve that follows the approximate shape of the sound's frequency profile curve. Using this visual aid you can go further by designing the equaliser backwards, that is, instead of drawing the correction you want you draw the bias you want to remove.

After loading the sound, duplicate it (menu Layers > Duplicate Layer) and turn it into curves (menu Operations > Horizontal Curves) that represent its frequency profile as shown in the image below : You can also use the sound's own frequency profile as a visual reference. Once done drawing the curve, fill it up (menu Operations > Fill from Maxima to Left), average it (menu Operations > Horizontal Average) then finally set the layer's blending mode to multiply (menu Layers > Multiplication).ДЇor best results the use of the lossless mode is recommended. Knowing this and using the editing tools available in the program we can first create an equalisation curve and use it to filter a sound.Д¬reate a new layer (menu Layers > New Layer) then select the White paint spray tool and start drawing a vertical curve, knowing that the left edge of the image represent 0 (complete silencing of that frequency) while the right edge represents the maximum amplification. This works by multiplying the respective intensities in frequency and time of each layer. The use of the lossless mode isn't recommanded when the lossless processing algorithm cannot be expected to give a satisfactory result.ДЇiltering in Photosounder can simply be done by multiplying (menu Layers > Multiplication) two layers together, one of the layers being the sound to be filtered, the other one being the filter. When the use of the lossless mode is recommended it is only when the effect is applied to a loaded sound. Generally the sound on which the effect will be applied will be known as Layer 0, the layers above being created and used to apply the desired effect. The list below details a few ways to reproduce classical audio processing effects using Photosounder's editing tools.

Photosounder freeware