Fractal dimension and iterated function system (IFS) for speech recognition

A new phoneme recognition method is presented which is based on fractal theory. The fractal dimension of the associated waveform and the IFS parameters that represent the waveform are used as the parameters for recognition.