Real-time speech signal analysis-synthesis of the Hoyer-Narris model using networked computers

The Hoyer-Harris model (H/sup 2/M) extracts the fundamental information of the speech signal to provide the knowledge for speech signal processing. Because of the complex algorithm, real-time analysis is not possible on a personal computer. The parallel processing design using networked computers, is used to provide the "real-time" responses of the H/sup 2/M model.