Self-organizing speech recognition for information extraction

摘要

A system is provided that trains a speech processing device to recognize any spoken language. The speech processing device is trained by receiving a group of speech frames ( 302 ). The group of speech frames is grouped into a group of segments ( 304 ). A model of each of the segments is created ( 306 ) and the segments are compared ( 308 ). The segments are grouped into clusters based on comparisons of the segments ( 310 ). A model for each of the clusters is built ( 312 ). The built cluster models enable the labeling of segments from speech in the language. Labeled segments (i.e., recognizer output) may be used to train topic classifiers. Speech topics may be classified on the basis of the generated labeled segments.