On creating reference templates for speaker‐independent recognition of words
暂无分享,去创建一个
The three aspects of a statistical approach to a pattern recognition problem are the selection of features, choice of a measure of similarity, and a method for creating the reference templates (patterns) used in the statistical tests. In this talk, we concentrate on methods for creating the reference templates. In particular, a method of combining word patterns from a number of speakers is proposed. The creation of templates, based on this method, is discussed. To test the ideas proposed here, a 54‐word vocabulary word recognition system was implemented. Using a selected set of the training words to create the template, recognition accuracies of 85% were obtained in a forced‐choice recognition test on the 54‐word vocabulary using eight new speakers. The correct word was within the top five choices 98% of the time. Using a strategy in which all the training words were used to create the templates, the recognition accuracy fell to 77% and the correct word was within the top five choices only 89% of the time.