Methods and apparatus for non-acoustic speech characterization and recognition

摘要: By simultaneously recording EM wave reflections (21) and acoustic speech information (24), the positions velocities of organs (2, 3, 4) as is articulated can be defined for each unit (20). Well time frames feature vectors (6, 7, 8, 9) describing speech, to degree required, formed. Such uniquely characterize (20) being frame. The onset rejection external noise, vocalized pitch periods, articulator conditions, accurate timing, identification speaker, recognition, organ mechanical parameters determined.