Accepted Papers

  • Detection of Automatic the VOT Value for Voiced Stop Sounds in Modern Standard Arabic (MSA)
    Sulaiman S. AlDahri, King Abdulaziz City for Science and Technology, Saudi Arabia
    Signal processing in current days is under studying. One of these studies focuses on speech processing. Speech signal have many important features. One of them is Voice Onset Time (VOT). This feature only appears in stop sounds. The human auditory system can utilize the VOT to differentiate between voiced and unvoiced stops like /p/ and /b/ in the English language. By VOT feature we can classify and detect languages and dialects. The main reason behind choosing this subject is that the researches in analyzing Arabic language in this field are not enough and automatic detection of VOT value in Modern Standard Arabic (MSA) is a new idea. In this paper, we will focus on designing an algorithm that will be used to detect the VOT value in MSA language automatically depending on the power signal. We apply this algorithm only on the voiced stop sounds /b/, /d/ and /d?/, and compare that VOT values automatically generated by the algorithm with the manual values calculated by reading the spectrogram. We created the corpus, and used CV-CV-CV format for each word, the target stop consonant is in the middle of word. The algorithm resulted in a high accuracy, and the error rate was 0.80%, 26.62% and 11.71% for the three stop voiced sounds /d/, /d?/ and /b/ respectively . The standard deviation was low in /d/ sound because it is easy to pronounce, and high in /d?/ sound because it is unique and difficult to pronounce.
  • Approaches in Using Expectation-Maximization Algorithm For Maximum Likelihood Estimation of The Parameters of A Constrained State Space Model With An External Input Series
    Chengliang Huang1,Xiao-Ping Zhang1 and Fang Wang2, 1Ryerson University Toronto,Canada and 2Wilfrid Laurier University Waterloo,Canada
    EM algorithm is popular in maximum likelihood estimation of parameters for state-space models. However, extant approaches for the realization of EM algorithm are still not able to fulfill the task of identification systems, which have external inputs and constrained parameters. In this paper, we propose new approaches for both initial guessing and MLE of the parameters of a constrained state-space model with an external input. Using weighted least square for the initial guess and the partial differentiation of the joint log-likelihood function for the EM algorithm, we estimate the parameters and compare the estimated values with the "actual" values, which are set to generate simulation data. Moreover, asymptotic variances of the estimated parameters are calculated when the sample size is large, while statistics of the estimated parameters are obtained through bootstrapping when the sample size issmall. The results demonstrate that the estimated values are close to the "actual" values.Consequently, our approaches are promising and can applied in future research.
  • A Novel ENF Extraction Approach for Region-of-Recording Identification of Media Recordings
    Majed El Helou, Abdel Wahab Turkmani, Rawan Chanouha and Samer Charbaji, American University of Beirut, Lebanon
    The electric network frequency (ENF) of power lines leaves its trace in nearby media recordings. The ENF signals vary in a consistent way in a given power grid. Therefore, it is possible to develop signal processing and machine learning techniques to identify the grid of origin by extracting attributes of the embedded ENF signal in recorded audio. This paper presents a model based on a novel ENF extraction technique with training on audio and power recordings from different grids. The proposed approach is based on correcting erroneously selected peaks from the Short Time Fourier Transform (STFT) by leveraging time correlations. These peaks are mistakenly taken for the frequency component belonging to the embedded ENF signal of the power grid and are corrected by the algorithm. Results on a test set of 50 recordings from nine different locations demonstrate the effectiveness of the proposed approach with an overall accuracy of 88%.
  • Automatic Translation of Arabic Sign To Arabic Text (ATASAT) System
    Abdelmoty M.Ahmed, Reda Abo Alez, Gamal Tharwat and Muhammad Taha, Al-Azhar University, Egypt
    Sign language continues to be the preferred tool of communication between the deaf and the hearing-impaired. It is a well-structured code by hand gesture, where every gesture has a specific meaning. In this paper has goal to develop a system for automatic translation of Arabic Sign Language. To Arabic Text (ATASAT) System this system is acts as a translator among deaf and dumb with normal people to enhance their communication.

    The proposed System consists of five main stages Video and Images capture, Video and images processing, Hand Signs Construction, Classification finally Text transformation and interpretation. This system depends on building a two datasets image features for Arabic sign language gestures alphabets from two resources: Arabic Sign Language dictionary and gestures from different signer's human, also using gesture recognition techniques, which allows the user to interact with the outside world. This system offers a novel technique of hand detection is proposed which detect and extract hand gestures of Arabic Sign from Image or video.

    In this paper we use a set of appropriate features in step hand sign construction and classification of based on different classification algorithms such as KNN, MLP, C4.5, VFI and SMO and compare these results to get better classifier.