Apparatus and method for classifying voiced, unvoiced and silent sections using neural network

摘要

1. TECHNICAL FIELD OF THE INVENTION;The present invention relates to an apparatus and a method for classifying voiced sounds, unvoiced sounds, and silent sections using neural networks.;2. The technical problem to be solved by the invention;An object of the present invention is to provide a method and apparatus for classifying voiced sounds, unvoiced sounds, and silenced sections by determining whether voiced sounds, unvoiced sounds, and silence are framed by using a predictive recursive neural network having feature parameters as inputs.;3. Summary of Solution of the Invention;The present invention provides signal input means for converting speech into a digital signal, feature parameter extracting means for extracting feature parameters, determination means for determining whether a frame to be analyzed is voiced, unvoiced or silent, and decision information. And voiced / unvoiced / silent information output means for outputting each time frame or outputting the same from the beginning to the end of the input.;4. Important uses of the invention;The present invention is used in a speech recognition device, speech synthesis device, speech analysis device.