ON THE NEURAL MODELING OF SPEECH PROCESSES.
Abstract
The work concerns the neural modeling of the human processes of speech production and perception. Part of this study deals with a new technique of speech synthesis. Newly defined generalized parameter waveforms, produced by an electronic waveform generator, are employed to approximate acoustic parameter transitions in the synthesis of dyad segments. The study also employs computer simulation studies to investigate neural models of: (1) the speech recognition process; (2) the control of speech production at the neural command level; and (3) feedback systems in the speech acquisition processes of babbling and mimicking. One of the preprocessing systems for speech recognition employs a new technique of detecting time sequences of features. In modeling these processes emphasis was on adaptive networks. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 15, 1966
- Accession Number
- AD0649160
Entities
People
- Charles C. Tappert
Organizations
- Cornell University