Pushed to cannam/sml-fft
A Fast Fourier Transform implementation in Standard ML. https://bitbucket.org/cannam/sml-fft Intended to be reasonably fast rather than beautiful -- it's largely imperative using lots of mutable state. Includes complex-complex and real-complex FFTs. Supports power-of-two sizes only. No dependencies apart from the Basis library. The complex-complex FFT is modelled on Nayuki's multi-language version at http://www.nayuki.io/page/free-small-fft-in-multiple-languages. The real-complex wrapper is modelled on KissFFT by Mark Borgerding. Chris Cannam, firstname.lastname@example.org. Free BSD-style licence, see COPYING for details.