Lexical stress detection on stress-minimal word pairs

The authors present a study on the use of lexical stress classification to aid in the recognition of phonetically similar words. They use a simple pattern recognition approach to determine which syllable is lexically stressed for phonetically similar word pairs (e.g., PERfect, perFECT) extracted from continuously spoken sentences. They use a combination of two features from the acoustic correlates of lexical stress, and assume multivariate Gaussian distributions to form a Bayesian classifier. The features used are normalized energy and duration of the vowel for each syllable of the word. They evaluate several normalization methods. Two sets of sentences were designed for this study. For the pilot experiment, the classification accuracy on words from the natural sentence set was 89.9% and on words from the control sentence set was 100%. To improve the performance, three-feature classifiers, which included two normalized energy features and one normalized duration feature, were developed. The classification accuracy on words from the natural sentence set was 97.23%.