A novel method of speech recognition using feedforward neural network technology

2002 ◽  
Author(s):  
J. Ahmad ◽  
H.A. Fatmi
Keyword(s):  
Neural Network ◽  
Speech Recognition ◽  
Feedforward Neural Network ◽  
Network Technology ◽  
Novel Method ◽  
Neural Network Technology

Automatic Image and Speech Recognition Based on Neural Network

2010 ◽  
Vol 3 (2) ◽  
pp. 1-17 ◽  
Author(s):  
Dariusz Król ◽  
Boguslaw Szlachetko
Keyword(s):  
Neural Network ◽  
Speech Recognition ◽  
Recognition System ◽  
License Plate ◽  
Network Technology ◽  
License Plate Recognition ◽  
Neural Structure ◽  
Automatic Speech Recognition System ◽  
System A ◽  
Neural Network Technology

The objective of this paper is to present a real-time mechanism for recognition of different objects using Spatiognitron neural network technology. Spatiognitron is based on a biological neural structure and the theory described in this paper presents what are known as Time Delay Neural Networks (TDNN). These are fields which enable the recognition of different features in the input object. The approach was verified by qualitative recognition process tests in commercial car license plate recognition using a NeuroCar based system. A second set of tests was carried out in a laboratory environment using NeuroScope, an automatic speech recognition system.

scholarly journals Novel sensitivity method for evaluating the first derivative of the feed-forward neural network outputs

2021 ◽  
Vol 8 (1) ◽  
Author(s):  
Ravi Kiran ◽  
Dayakar L. Naik
Keyword(s):  
Neural Network ◽  
Feedforward Neural Network ◽  
Analytical Quality ◽  
Feed Forward Neural Network ◽  
First Derivative ◽  
Network Output ◽  
Novel Method ◽  
Classification Tasks ◽  
Trained Neural Network ◽  
Sensitivity Method

AbstractEvaluating the exact first derivative of a feedforward neural network (FFNN) output with respect to the input feature is pivotal for performing the sensitivity analysis of the trained neural network with respect to the inputs. In this paper, a novel method is presented that computes the analytical quality first derivative of a trained feedforward neural network output with respect to the input features without the need for backpropagation. To this end, the complex step derivative approximation is illustrated, and its implementation in the framework of the feedforward neural network is described. Artificial datasets are generated, and the efficacy of the proposed method for both regression and classification tasks is demonstrated. The results obtained for the regression task indicated that the proposed method is capable of obtaining analytical quality derivatives, and in the case of the classification task, the least relevant features could be identified.

scholarly journals UNORGANIZED AIR RELEASES WITH RADIOACTIVE AEROSOLS FROM THE NEW SAFE CONFINIMENT OF CHNPP INTO THE ENVIRONMENT

2021 ◽  
pp. 181-186
Author(s):  
P.G. Krukovskyi ◽  
Ye.V. Diadiushko ◽  
D.J. Skliarenko ◽  
I.S. Starovit
Keyword(s):  
Neural Network ◽  
Computational Model ◽  
Network Technology ◽  
Building Structures ◽  
Flow Rates ◽  
Radioactive Aerosols ◽  
Computational Performance ◽  
Lack Of Information ◽  
Chernobyl Npp ◽  
Neural Network Technology

The New Safe Confinement (NSC) of the Chernobyl NPP, which isolates the destroyed reactor and the “Shelter Object” from the environment, is not airtight, so the problem is the lack of information on the flow of unorganized air with radioactive aerosols outside the NSC. This work presents computational model of the hydraulic state of the NSC, which allows to determine these flow rates through the leaks in the shells and building structures under the walls of the NSC. In addition to the developed model, the NSC hydraulic state model, created by neural network technology, was tested, which showed similar results and much higher computational performance, which allows its use for analysis and prediction of NSC`s hydraulic state in real time.

Sign in / Sign up

Export Citation Format

Share Document