Learning Phonetic Features Using Connectionist Networks

A method for learning phonetic features from speech data using connectionist networks is described. A temporal flow model is introduced in which sampled speech data flows through a parallel network from input to output units. The network uses hidden units with recurrent links to capture spectral/temporal characteristics of phonetic features. A supervised learning algorithm is presented which performs gradient descent in weight space using a coarse approximation of the desired output as an evaluation function. A simple connectionist network with recurrent links was trained on a single instance of the word pair "no" and "go", and successful learned a discriminatory mechanism. The trained network also correctly discriminated 98% of 25 other tokens of each word by the same speaker. A single integrated spectral feature was formed without segmentation of the input, and without a direct comparison of the two items.