Although, speech recognition systems are widely used and their accuracies are continuously increased, there is a considerable performance gap between their accuracies and human recognition ability. This is partially due to high speaker variations in speech signal. Deep neural networks are among the best tools for acoustic modeling. Recently, using hybrid deep neural network and hidden Markov model (HMM) leads to considerable performance achievement in speech recognition problem because deep networks model complex correlations between features. The main aim of this paper is to achieve a better acoustic modeling by changing the structure of deep convolutional neural network (CNN) in order to adapt speaking variations. In this way, existing models and corresponding inference task have been improved and extended. Here, we propose Adaptive windows convolutional neural network (AWCNN) to analyze joint temporal-spectral features variation. AWCNN changes the structure of CNN and estimates the probabilities of HMM states. We propose Adaptive windows convolutional neural network in order to make the model more robust against the speech signal variations for both single speaker and among various speakers. This model can better model speech signals. The AWCNN method applies to the speech spectrogram and models time-frequency varieties. This network handles speaker feature variations, speech signal varieties, and variations in phone duration. The obtained results and analysis on FARSDAT and TIMIT datasets show that, for phone recognition task, the proposed structure achieves 1. 2%, 1. 1% absolute error reduction with respect to CNN models respectively, which is a considerable improvement in this problem. Based on the results obtained by the conducted experiments, we conclude that the use of speaker information is very beneficial for recognition accuracy.

Deep convolutional neural networks (CNNs) have attained remarkable success in numerous visual recognition tasks. There are two challenges when adopting CNNs in real-world applications: a) Existing CNNs are computationally expensive and memory intensive, impeding their use in edge computing; b) there is no standard methodology for designing the CNN architecture for the intended problem. network pruning/compression has emerged as a research direction to address the first challenge, and it has proven to moderate CNN computational load successfully. For the second challenge, various evolutionary algorithms have been proposed thus far. The algorithm proposed in this paper can be viewed as a solution to both challenges. Instead of using constant predefined criteria to evaluate the filters of CNN layers, the proposed algorithm establishes evaluation criteria in online manner during network training based on the combination of each filter’s profit in its layer and the next layer. In addition, the novel method suggested that it inserts new filters into the CNN layers. The proposed algorithm is not simply a pruning strategy but determines the optimal number of filters. Training on multiple CNN architectures allows us to demonstrate the efficacy of our approach empirically. Compared to current pruning algorithms, our algorithm yields a network with a remarkable prune ratio and accuracy. Despite the relatively high computational cost of an epoch in the proposed algorithm in pruning, altogether it achieves the resultant network faster than other algorithms.

convolutional neural network is one of the effective methods for classifying images that performs learning using convolutional, pooling and fully-connected layers. All kinds of noise disrupt the operation of this network. Noise images reduce classification accuracy and increase convolutional neural network training time. Noise is an unwanted signal that destroys the original signal. Noise changes the output values of a system, just as the value recorded in the output differs from its actual value. In the process of image encoding and transmission, when the image is passed through noisy transmission channel, the impulse noise with positive and negative pulses causes the image to be destroyed. A positive pulse in the form of white and a negative pulse in the form of black affect the image. The purpose of this paper is to introduce dynamic pooling which make the convolutional neural network stronger against the noisy image. The proposed method classifies noise images by weighting the values in the dynamic pooling region. In this research, a new method for modifying the pooling operator is presented in order to increase the accuracy of convolutional neural network in noise image classification. To remove noise in the dynamic pooling layer, it is sufficient to prevent the noise pixel processing by the dynamic pooling operator. Preventing noise pixel processing in the dynamic pooling layer prevents selecting the amount of noise to be applied to subsequent CNN layers. This increases the accuracy of the classification. There is a possibility of destroying the pixels of the entire window in the image. Due to the fact that the dynamic pooling operator is repeated several times in the layers of the convolutional neural network, the proposed method for merging noise pixels can be used many times. In the proposed dynamic pooling layer, pixels with a probability of p being destroyed by noise are not included in the dynamic pooling operation with the same probability. In other words, the participation of a pixel in the dynamic pooling layer depends on the health of that pixel value. If a pixel is likely to be noisy, it will not be processed in the proposed dynamic pooling layer with the same probability. To compare the proposed method, the trained VGG-Net model with medium and slow architecture has been used. Five convolutional layers and three fully connected layers are the components of the proposed model. The proposed method with 26% error for images corrupted with impulse noise with a density of 5% has a better performance than the compared methods. Increased efficiency and speed of convolutional neural network based on dynamic pooling layer modification for noise image classification is seen in the simulation results.

Speaker recognition is a process of recognizing persons based on their voice which is widely used in many applications. Although many researches have been performed in this domain, there are some challenges that have not been addressed yet. In this research, Neutrosophic (NS) theory and convolutional neural networks (CNN) are used to improve the accuracy of speaker recognition systems. To do this, at first, the spectrogram of the signal is created from the speech signal and then transferred to the NS domain. In the next step, the alpha correction operator is applied repeatedly until reaching constant entropy in subsequent iterations. Finally, a convolutional neural networks architecture is proposed to classify spectrograms in the NS domain. Two datasets TIMIT and Aurora2 are used to evaluate the effectiveness of the proposed method. The precision of the proposed method on two datasets TIMIT and Aurora2 are 93.79% and 95.24%, respectively, demonstrating that the proposed model outperforms competitive models.

Finger-Knuckle-Print (FKP) is an accurate and reliable biometric in compare to other hand-based biometrics like fingerprint because of the finger's dorsal region is not exposed to surfaces. In this paper, a simple end-to-end method based on convolutional neural network (CNN) is proposed for FKP recognition. The proposed model is composed only of three convolutional layers and two fully connected layers. The number of trainable parameters hereby has significantly reduced. Additionally, a straightforward method is utilized for data augmentation in this paper. The performance of the proposed network is evaluated on Poly-U FKP dataset based on 10-fold cross-validation. The best recognition accuracy, mean accuracy and standard deviation are 99.83%, 99.18%, and 0.76, respectively. Experimental results show that the proposed method outperforms the state-of-the-arts in terms of recognition accuracy and the number of trainable parameters. Also, in compare to four fine-tuned CNN models including AlexNet, VGG16, ResNet34, and GoogleNet, the proposed simple method achieved higher performance in terms of recognition accuracy and the numbers of trainable parameters and training time.

Due to the cursive-ness and high variability of Persian scripts, the segmentation of handwritten words into sub-words is still a challenging task. These issues could be addressed in a holistic approach by sidestepping segmentation at the character level. In this paper, an end-to-end holistic method based on deep convolutional neural network is proposed to recognize off-line Persian handwritten words. The proposed model uses only five convolutional layers and two fully connected layers for classifying word images effectively, which can lead to a substantial reduction in parameters. The effect of various pooling strategies is also investigated in this paper. The primary goal of this article is to ignore handcrafted feature extraction and attain a generalized and stable word recognition system. The presented model is assessed using two famous handwritten Persian word databases called Sadri and IRANSHAHR. The recognition accuracies were obtained at 98.6% and 94.6%, on Sadri and IRANSHAHR datasets respectively, and outperformed the state-of-the-art methods.