LVQ-based speech recognition with high-dimensional context vectors

In this paper we have applied the Learning Vector Quantiza-tion methods, including the latest developments 1, 3, 2, 4], to the task of Finnish speaker-dependent speech recognition. The main objective was to study the eeect of radically increasing the dimensionality of the context vectors. The high-dimensional feature vectors in our work represent the whole phoneme and they are formed by both averaging and concatenating short-time feature vectors within a time domain window. Excellent results are achieved in separate phoneme classiication of Finnish speech. Moreover, we also show how this method can be applied in combined labeling and segmentation of continuos speech. In this task we use an additional segmentation LVQ-codebook, and combine the information using HMMs.