Speech Recognition Using Modified General Fuzzy Min-Max Neural Network
نویسندگان
چکیده
In this paper, we report the results of Marathi (Language spoken in the state of Maharashtra, India) spoken digit recognition using General Fuzzy Min-Max Neural Network (GFMM NN)[1] and Modified General Fuzzy MinMax Neural Network (MGFMM NN), which is obtained by modifying the transfer function of output layer of GFMM NN.
منابع مشابه
Object Recognition Using Reflex Fuzzy Min-Max Neural Network with Floating Neurons
This paper proposes an object recognition system that is invariant to rotation, translation and scale and can be trained under partial supervision. The system is divided into two sections namely, feature extraction and recognition sections. Feature extraction section uses proposed rotation, translation and scale invariant features. Recognition section consists of a novel Reflex Fuzzy MinMax Neu...
متن کاملColor Object Recognition Using General Fuzzy Min Max Neural Network
A hybrid approach based on Fuzzy Logic and neural networks with the combination of the classic Hu & Zernike moments joined with Geodesic descriptors is used to keep the maximum amount of information that are given by the color of the image. These moments are calculated for each color level and geodesic descriptors are applied directly to binary images to get information about the general shape ...
متن کاملPersian Printed Numeral Characters Recognition Using Geometrical Central Moments and Fuzzy Min-Max Neural Network
In this paper, a new proposed system for Persian printed numeral characters recognition with emphasis on representation and recognition stages is introduced. For the first time, in Persian optical character recognition, geometrical central moments as character image descriptor and fuzzy min-max neural network for Persian numeral character recognition has been used. Set of different experiments ...
متن کاملApplication of the Fuzzy Min-max Neural Network Classifier to Problems with Continuous and Discrete Attributes
The fuzzy min-max classiication network constitutes a promisimg pattern recognition approach that is based on hy-berbox fuzzy sets and can be incrementally trained requiring only one pass through the training set. The deenition and operation of the model considers only attributes assuming continuous values. Therefore, the application of the fuzzy min-max network to a problem with continous and ...
متن کاملEnhancing Performance of Relational Fuzzy Neural Networks with Square BK-Products
In this paper, we extend research done in max-min fuzzy neural networks in several important ways. We replace max and min operations used in the fuzzy operations by more general t-norms and co-norms, respectively. In addition, instead of the Łukasiewicz equivalence connective used in network of ReyesGarcia and Bandler, we employ in our hybridization a variety of equivalence connectives. We expl...
متن کامل