Text-independent talker identification system combining connectionist and conventional models

Several techniques have been used for speaker identification which have different characteristics and capabilities. The respective merits of three different systems respectively employing neural networks, hidden Markov models, and multivariate autoregressive models are compared. A novel text-independent speaker identification system based on the cooperation of these different techniques is presented. This system outperforms previous models and can handle a large number of speakers. It is argued that modular architectures present significant advantages, such as their learning speed, their generalization and representation capabilities, and their ability to satisfy constraints imposed by hardware limitations