A Neural Network Clustering Technique for Text-independent Speaker

A clustering algorithm for speaker identification based on neural networks is described. This technique is modeled after a previously developed technique in which an N-way speaker identification task is partitioned into N*(N-1)/2 two-way classification tasks. Each two-way classification task is performed using a small size neural network which is a two-way, or pair-wise, network. The decisions of these two-way networks are then combined to make the N-way speaker identification decision (Rudasi and Zahorian, 1991 and 1992). Although very accurate, this method has the drawback of requiring a very large number of pair-wise networks. In the new approach two-way neural network classifiers, each of which is trained only to separate two speakers, are also used to separate other pairs of speakers. Thus, in effect, speakers are clustered according to each pair-wise classifier. This method is able to greatly reduce the number of pair-wise classifiers required for making an N-way classification decision, especially when the number of speakers is very large. For 100 speakers extracted from TIMIT database, we were able to reduce the required number of pair-wise classifiers by a factor of 5, with no degradation in performance when 2 seconds or more of speech are used for identification. We obtained 100% text-independent speaker identification accuracy for 200 speakers with approximately 6 seconds of speech from each speaker and 97% when 2 seconds of speech were used.