Speech Recognition using Epochwise Back Propagation

wrangle apprehension exploitation eonwise Back PropagationInternational Journal of Computer Applications (0975 8887) Volume * nary(prenominal)*, ___________ 2013Speech Recognition development the Epochwise Back Propagation by style of condemnation AlgorithmNeelima Rajput S.K.VermaDepartment of C.S.E. Department of C.S.E.GBPEC, Pauri Gharwal, GBPEC, Pauri Gharwal,Uttrakhand, India. Uttrakhand, IndiaABSTRACTIn this paper, artificial anxious meshworks argon employ to accomplish the English first rudiment quarrel recognition. The ambitious issue in the atomic number 18a of barbarism recognition is to design an faithful and effective tongue recognition dust. We implemented a sore selective information classification rule, where we use skittish lootworks, which argon turn backed using the careful term determine of audio target. This method gave comparable result to the already implemented anxious meshworks. In this paper, Back denotation uneasy network calculating machine architecture used to recognize the date varying infix information, and provides correct accu regularise results for the English Alphabet linguistic process recognition. The Epochwise Back Propagation by time (BPTT) algorithmic rule uses the eon stupefy of remark bode to control the network structures and yields the satisfactory results.Key enunciates Artificial flyaway net profit, Back Propagation queasy lucre, Epoch, Speech Recognition.I INTRODUCTIONSpeech recognition establishment enables the forge to project the human actors line and react accordingly. It on the whole toldows the machine to automatic each(prenominal)y understand the human spoken utterances with the speech signal inter-group communication on and pattern recognition. In this approach is the machine converts the voice signal into the suitable text or command through the process of identification and understanding. Speech recognition is emerges as a vast technology in cu rrent time. It also plays an grievous role in information theory, acoustics, phonetics, linguistics, and pattern recognition theory and neurobiology disciplines. speech recognition technology become a key technology in the computer information bear upon technology as there is speedy advancement in the software, hardware and information technology. The features of remark audio signal are compared with the voice template stored in the computer database in speech recognition st posegy by using the computer systems. Recognition results are mainly depends upon the matching techniques used for matching the audio signal characteristics. To ameliorate the recognition rate and fail recognition results spooky networks are used.A queasy network is a powerful tool which used to alter and represent the complicated input end products. unquiet nets are underlying aloney co-ordinated networks of relatively simple processing unit of measurements, or nodes that work simultaneously. They are designed to mimic the function of human neuron biological networks. The processing units of neural networks are termed as neurons. A neural network provides better results every(prenominal) over the living approaches in speech recognition systems 1.2 BASICS OF aflutter NET spirtSThe basics of neural networks are discussed as follows. There are some(prenominal) different types of neural networks, unless they all have four basic and common attributes affect units Connections Computing procedure raising procedure.Processing UnitsA neural network contains several processing units, which are just ab come to the fore analogous to neurons in the human brain. All these units activate in couple and perform the task simultaneously. Processing units are responsible for the overall numeration there is no any other unit for the corporation of their activity. separately processing unit computes a scalar function of all its topical anaesthetic inputs at every moment of time and consequently further dissipate the result to their neighboring units 2. The units in a neural network are basically classified into input units, which used to receive data from the away(p) recondite units, used to internally transform the data and output units, which service decisions or target signals.2.2 ConnectionsAll processing units in a neural network are organized in to a defined network topology by a dress of connections, or weights, shown as lines in a diagram. all(prenominal) weight consist a real value, which ranging from to + . The value of a weight represents how much impact a unit has on its neighbor units a positive weight causes one unit to excite another, piece a negative weight causes one unit to inhibit another. Weights of the processing units are normally one-directional (from input units towards output units), but it may be two-directional sometimes, especially when there is no distinction amid input and output units.Figure 1.Neural network topologie s (a) unstructured, (b) layered, (c) recurrent, (d) modular 3.The above figure shows the topology of different neural networks. Unstructured Neural Network is mainly used in processor which is operated in parallel to provide the deliberational power for the speech recognition system. Layered neural net algorithms are self-organizing and their internal speech model maximizes the military operation and stimulates better results comparable to existing systems.Recurrent Neural network are mainly used in pattern sequencing as it follows the sequences changes time to time. Modular neural network are used in designing of complex systems by using simple components.Neural nets set up an approach to computation that mimics biological nervous systems. Algorithms base on neural nets have been proposed to conduceress speech recognition tasks which humans perform with lower-ranking apparent effort.2.3 Computation ProcedureComputation of neural networks begins after applying an input data t o the input units of the network. Then the energizing function of all the units is reckon either simultaneously or independently depends upon the structure of the neural network. The computation process in the unstructured network is termed as spreading activating and in layered network is known as forward file name extension as it proceed from the input units to the output units. First we compute the net input of the processing unit and and so we compute the output activation function of the net input to update a accustomed processing unit.2.4 Training ProcedureTraining a network means enable the connections adaptive so that the network shows the actual computational behavior for all the input patterns. In training process usually weights are updated but sometimes the modification of network topology also takes place, i.e., addition and ablation of connections from the network topology 4. Modification of weight is easy and beneficial than topology modification as a network wi th bulk connections able to set any of its weight zero, which is equivalent as deleting such weights.3. RELATED WORKThe past research concluded that the use of the neural networks in the speech recognition system provides the better recognition result compared to the other existing approaches. The latest field of neural networks actually started in the 19th century, when neurobiologists prototypic introduce the extensive studies of the human nervous system 5. Cajal (1892) determined that the nervous system is comprised of some basic units i.e. discrete neurons, which communicate with the other neurons by sending electrical signals down their long axons, which ultimately activated and touch the dendrites(receptive areas) of thousands of other neurons, transmitting the electrical signals through synapses (points of contact, with variable resistance). Firstly, the different kinds of neurons were identified, and then analyze their electrical responses, and finally their patterns of co nnectivity and the brainsgross functional areas were mapped out. According to the neurobiologists study the functionality of individual neurons are quite simple and easy. Whereas to determine how neurons worked together to pass high level functionality, such as perception and cognition are very difficult.6In 1943 McCulloch and Pitts proposed the first enumeration model of a neuron, named as binary threshold unit, whose output was either 0 or 1 depending on whether its net input exceeded a given threshold. There are various approaches proposed by the researchers to design an accurate speech recognition system for various purposes. In 7 Al-Alaoui algorithm is used to train the neural network. This method gives the comparable better results to the already implemented recondite markov model (HMM) for the recognition of the words. This algorithm also overcomes the disadvantages of the HMM in the recognition of sentences. An algorithm based on neural network classifier 8 for speech reco gnition used a new Viterbi net architecture which is recognized the input patterns and provided an accuracy of recognition rate more than 99% on a large speech database. This system is used for isolated word recognizer. In 9 source accomplishes the isolated word speech recognition using the neural network. The methodology of this approach is to invoke the feature of speech signals using the Digital Signal Processing techniques and then classification using the Artificial Neural Network. This algorithm concluded that the better accurate recognition results are obtained from the probalistic Neural Network PNN. In 10 author implemented a pre- trained deep neural network using the hidden markov model (DNN-HMM) hybrid architecture which is used to train the DNN to produce the better recognition results of large vocabulary speech database.4. PROPOSED WORKSpeech recognition using the Epochwise Back annex through time algorithm is proposed in this paper.In the proposed system neural netw ork training is based on the calculation of epoch of the audio signal and then used these epoch value for the training of the neural network. The input data sets used to train the neural network bottomland be partitioned in to the independent epochs. Each epoch representing a temporal value of the input data. Back propagationneural network used in the system in following steps.First have and fix the architecture for the network, which will contain input, hidden and output units, all of which will contain sigmoid functions.Assign the weights between all the nodes. The assignments of weights usually between -0.5 and 0.5.Each training example is used, one after another, to re-train the weights in the network.After calculating from each one epoch for input audio data, a termination condition is checked.In neural network architecture the weights of input and hidden layers are adjusted according to the target output values 11 The input data is considered as E which is propagated throug h the network so that we can record all the observed values Oi(E) for the output nodes Oi. At the same time, we record all the observed values hi(E) for the hidden nodes. Then, for each output unit Ok, we calculate its epoch as follows(1)The epoch harm from the output units are used to calculate epoch ground for the hidden units. In fact, this method gets its name because we propagate this information backwards through the network 12. For each hidden unit Hk, we calculate the epoch as follows(2)Here, we take the epoch term for every output unit and regurgitate it by the weight from hidden unit Hk to the output unit. We then add all these together and multiply the sum by hk(E)*(1 hk(E)). Having work out all the epoch values associated with each unit (hidden and output), we can now murder this information into the weight changes ij between units i and j. The calculation is as follows for weights wij between input unit Ii and hidden unit Hj, and summation of all units are as(3)Ba ck Propagation Neural Network architecture is shown in below figureFig.2 Back propagation Neural NetworkThe main steps of this system are defined as follows. represent the input audio Signal.Extract the epoch valuesTrain the neural network on the al-Qaida of epoch values.Applied the back propagation neural network for the classification.Matching the input data with the trained data. recognise the input.The data flow diagram of the proposed system represented in given figureCFigure 2 Proposed approaches for Speech Recognition5. experimental RESULTSThe experiment conducted on the audio database of English alphabets. Matlab R2010a (Math works) in Windows 7 was used to implement the proposed algorithm. The input signal is used to calculate the epoch values and then by using calculated epoch values the neural network is trained.The epoch rate of multi-layered networks over a training set could be calculated as the number of miss-classified data values. There are many output nodes, all of which could potentially misfire (e.g., giving a value close to 1 when it should have output 0 and vice-versa), we can be more sophisticated in our epoch evaluation. In radiation pattern the overall network epoch is calculated asThis is not as complicated as it first appears. The calculation simply involves working out the difference between the observed output for each output unit and the target output and squaring this to make sure it is positive, then adding up all these squared differences for each output unit and for each input signal. The calculated epoch values for audio signal A is listed in tabular form.Table 1 Calculated Epoch values for input dataThen epoch values are basically instant of noteworthy excitation of the vocal-tract system during production of speech. Table 1 represented the Epoch values for the input English alphabet A in terms of time in mili seconds. These epoch values are stored in the computer for the further matching by using the neural networks to the recognition purpose. Similarly, epoch values are calculated for each input English alphabet and then train the network through these input patterns and stored in the system for recognitionFigure 3 Performance graphs of epoch values.The figure 3 shows the performance of the system based on epoch values. The best organisation performance epoch value is selected from the different epoch on the basis of Mean Square error. The graph is plotted on epoch and means squared error values.Table 2 Performance of the proposed systemIn Table 2 the recognition rate is calculated for five input English alphabet which is 98.8%. The new proposed Epochwise Back propagation through time algorithm yields the satisfactory results.6. CONCLUSIONIn this paper, we implemented the Epochwise Back propagation through the time varying epoch calculation. The experiment is conducted on the humble set of English Language alphabet to calculate the recognition rate of the system. Some different sound samples ( i.e., with different sampling frequency) of each alphabet are taken and used for testing the system. The above results show the performance of our proposed algorithm in speech recognition.7. REFERENCESJianliang Meng, Junwei Zhang,Haoquan Zhao, Overview of the speech Recognition Technology, 2012 fourth part International Conference on Computational and Information Sciences.L. Fausset, Fundamentals of Neural Networks. PrenticeHall Inc., 1994, ch 4.Jiang Ming Hu, in the Yuan Baozong, Lin Biqin. Neural networks for speech recognition research and progress. Telecommunications Science, 1997, 13(7)1-6.H. Boulard and N. Morgan, unceasing speech recognition by connectionist statistical methods, IEEE Trans. Neural Netw., vol. 4, no. 6, pp. 893909, Nov. 1993.Mohamad Adnan Al-Alaoui, Lina Al-Kanj, jemmy Azar, and Elias Yaacoub, Speech recognition using Artificial Neural Network and confidential Markov Model, IEEE Multidisciplinary Engineering Education Magazine Vol. 3, No.3, September 2008. RICHARD P. LIPPMANN, Neural Network Classifiers for Speech Recognition The Lincoln Laboratory Journal, Volume 1, Number 1 (1988)Gulin Dede, Murat Husnu Sazl, Speech recognition using artificial neural network. Digital signal processing 2009 Elsevier Inc.George E. Dahl, Dong Yu, Li Deng, Alex Acero, Context-Dependent Pre-Trained Deep Neural Networks for Large-Vocabulary Speech Recognition , IEEE Transaction on Audio, and Language Processing, Vol. 20, No.1, January 2012.H. Paugam-Moisy, Parallel neural computing based on network duplicating,in Parallel Algorithms for Digital Image Processing, Computer resource and Neural Networks, ed., I. Pitas, 305340, JohnWiley, (1993).Stefano Scanzio, Sandro Cumani, Roberto Gemello, Franco Mana, P. Laface, Parallel implementation of Artificial Neural Network Training for Speech Recognition. Pattern recogonition letters, 2010 Elsevier B.V.N. Morgan and H. Bourlard, Continuous speech recognition using multilayer perceptrons with hidden Markov mode ls, in Proc. ICASSP,1990, pp. 413416.Y. Hifny and S. Renals, Speech recognition using augmented conditional haphazard fields, IEEE Trans. Audio, Speech, Lang. Process., vol.17, no. 2, pp. 354365, Feb. 2009.