| 摘要 |
A speech recognition capability in which words of spoken text are identified based on the contour of sound waves representing the spoken text. Variations in the contour of the sound waves are identified, features are assigned to those variations, and then the features are mapped to sound constructs to provide the words. |
