Informações Principais
     Resumo
     Abstract
     Introdução
     Conclusão
     Download
  
  
  
 
Abstract
 
 
Acadêmico(a): Arnoldo Uber Junior
Título: Classes Automation Ole para Reconhecimento de Fala Através da Rede Neural RBF-Fuzzy Artmap
 
Abstract:
This work has as objective facilitates the use of artificial neural network (ANN) applied to the speech recognition in the software development. It will be possible to the developer, to train ANN (RBF-Fuzzy Artmap model) will be used to workout, with the words that he wants. To generate a knowledge base resulting from the training of ANN and through an interface Automation OLE (Object Linking and Embedding), to incorporate this technology in their applications. The interface Automation OLE will work ANN when he needs to recognize a spoken word, tends as objective, the execution of a certain action. To demonstrate the use of the developed classes, it was specified as example, an application of speech recognition that uses them, making actions worked by the pronunciation of the numbers of 0 (zero) at 9 (nine) for the user.