The Fisher Transform changes the probability density function of any waveform so the transformed output has an approximately Gaussian.

In the case of the Fisher Transform indicator used an example by John F. Ehlers in his "Using The Fisher Transform" article in the November 2002 issue of Stocks & Commodities magazine, the Fisher Transform was applied a variation of a stochastic where the midpoint of the current bar was used instead of the current price in its calculation and the values run from -1 to 1 instead of 0 to 100.

The stochastic was just used to normalize prices, but applying the Fisher Transform to this modified stochastic has become what is considered the standard Fisher Transform indicator.

The idea is to transform price to minimize extreme movement and create timely clearly identifiable sharp turning points where the rate of change in price is the largest.

Read more on Fisher Transform at Investopedia.