IJENS Publisher Indexed In SCOPUS


Title: Speech to text Synthesis from Video Automated Subtitling Using Levinson Durbin Method of MEL Frequency Cepstrum Coefficient
Author(s): Sathish Kumar Selvaperumal, Vinesh Thiruchelvam, Chandrasekharan Nataraj, Willy Tong Chen Hung
Pages: 1-12 Paper ID:166305-1701-4343-IJVIPNS-IJENS Published: February, 2017
Abstract: Speech processing is relatively important since the past decade and to be researched by many more researchers in future. The objective of speech processing is to make sure the important information to be transferred from one to another by further enhancing the transferring method. Most of the things already digitalized and same will be applied into speech processing. In this paper, a Graphical User Interface (GUI) has been developed with the help of matlab, in order to extract the audio speech from the video under test. Noise has been removed by the designed Butterworth filter and by employing frame by frame analysis, words in the audio speech are separated using thresholding method. Feature extraction is done with the help of Levinson-Durbin method of Mel Frequency Cepstrum Coefficient as compared to the Linear Predictive Coding; and are given as input to the neural network for training and recognition. All of the spoken speech in the video are displayed into a text box using GUI. Finally, the system has achieved a recognition rate of 76% to 79% by recognizing 60 samples out of 150 samples as compared to the other researches which are having lesser samples. This work has been implemented into a new application which is video automated subtitling rather than a normal regular speech processing system
Keywords: Speech processing, Butterworth filter, Feature extraction, Levinson Durbin, Linear Predictive Coding, Neural network.
Full Text (.pdf)  International Journals Of Engineering and Science | 638 KB
Title: HD Remote Sensing Image Protection Approach based on Modified AES Algorithm
Author(s): Basher J. Hamza
Pages: 13-23 Paper ID:175101-8282-IJVIPNS-IJENS Published: February, 2017
Abstract: Sheltering of multimedia information from different types of adversary has become important for people and states. A high definition image has a large amount of data, and thus, keeping it secret is difficult. Another challenge that security algorithms must face with respect to high definition images in remote sensing applications is pattern appearances, which results from existing regions with high density in the same color, such as sky regions. A New image security algorithm collect between ciphering algorithm and hiding technique is suggested in this paper. The new hiding algorithm proposed here starts by applying reordering and scrambling operations to the six Most Significant Bit planes of the secret image, and then, it hides them in an unknown scene cover image using arithmetic operation. Then, Thestego-image encrypted using modified AES algorithm. Several parameters were used including the visual scene, histogram analysis, entropy, security analysis, and execution time to evaluate the proposed algorithm.
Keywords: Image processing. Image encryption.Image hiding.Hybrid security system.
Full Text (.pdf)  International Journals Of Engineering and Science | 1,082 KB