A neural network for synthesizing the pitch of an acoustic source

The authors describe a neural network model capable of generating a spatial representation of the pitch of an acoustic source that can be naturally imbedded in an architecture for signal separation. The model is a type of pattern matching model. The input to the pitch detecting module is a spectral representation, which is discussed. Pitch is one of several auditory percepts used by humans to separate multiple sound sources in the environment from each other. The model provides a neural instantiation of a type of harmonic sieve. It is capable of quantitatively simulating a large body of psychoacoustical data, including new data on octave shift perception