Voiced-unvoiced-silence classification of speech using neural nets

The authors describe a fast training algorithm for feedforward neural nets and apply it to a two-layer neural network to classify segments of speech as voiced, unvoiced, or silence. The speech classification method is based on features computed for each speech segment and used as input to the network. The network weights are trained using a novel fast training algorithm which uses a quasi-Newton error minimization method with a positive-definite approximation of the Hessian matrix. When used for voiced-unvoiced-silence classification of speech frames, the network performance compares favorably with that of current approaches