Nakamura, R. [UNESP]Pereira, L. [UNESP]Silva, D. [UNESP]Cardozo, P. [UNESP]Pereira, C. [UNESP]Ferasoli, H. [UNESP]Alves, S. [UNESP]Pires, R. [UNESP]Spadotto, A. [UNESP]Papa, J. [UNESP]2014-05-272014-05-272012-10-01INES 2012 - IEEE 16th International Conference on Intelligent Engineering Systems, Proceedings, p. 67-71.http://hdl.handle.net/11449/73611Voice-based user interfaces have been actively pursued aiming to help individuals with motor impairments, providing natural interfaces to communicate with machines. In this work, we have introduced a recent machine learning technique named Optimum-Path Forest (OPF) for voice-based robot interface, which has been demonstrated to be similar to the state-of-the-art pattern recognition techniques, but much faster. Experiments were conducted against Support Vector Machines, Neural Networks and a Bayesian classifier to show the OPF robustness. The proposed architecture provides high accuracy rates allied with low computational times. © 2012 IEEE.67-71engAccuracy rateBayesian classifierComputational timeMachine learning techniquesMotor impairmentsNatural interfacesOptimum-path forestsPattern recognition techniquesProposed architecturesRobot interfaceVoice interfacesLearning systemsPattern recognitionUser interfacesForestryInterfacesNetworksOptimizationPatternsRobotsTrabalho apresentado em evento10.1109/INES.2012.6249804Acesso aberto2-s2.0-84866652350