Speech recognition representing a communication between computers and human as a sub eld of computational linguistics or natural language processing has a long history. Automatic Speech Recognition (ASR), Text To Speech (TTS), speech to text, Continuous Speech Recognition (CSR), and interactive voice response systems are dierent approaches to solving problems in this area. The performance improvement is partially attributed to the ability of the Deep neural Network (DNN) to model complex correlations in speech features. In this paper, unlike the use of conventional model for sequential data like voice that employs Recurrent neural networks (RNNs) with the emergence of dierent architectures in deep networks and good performance of Conventional neural networks (CNNs) in image processing and feature extraction, the application of CNNs was developed in other domains. It was shown that prosodic features for Persian language could be extracted via CNNs for segmentation and labeling speech for short texts. By using 128 and 200 lters for CNN and special architectures, 19. 46 error in detection rate and better time consumption than RNNs were obtained. In addition, CNN simplies the learning procedure. Experimental results show that CNN networks can be a good feature extractor for speech recognition in various languages.

In the past three decades, the use of smart methods in medical diagnostic systems has attracted the attention of many researchers. However, no smart activity has been provided in the  eld of medical image processing for diagnosis of bladder cancer through cystoscopy images despite the high prevalence in the world. In this paper, two well-known convolutional neural networks (CNNs) and a multilayer neural network was applied to classify bladder cystoscopy images. One of the most im-portant issues in training phase of neural networks is determining the learning rate because selecting too small or large learning rate leads to slow convergence, volatility and divergence, respectively. Therefore, an algorithm is required to dynamically change the convergence rate. In this respect, an adaptive method was presented for determining the learning rate so that the multilayer neural network could be improved. In this method, the learning rate is determined using a coe cient based on the di erence between the accuracy of training and validation according to the output error. In addition, the rate of changes is updated according to the level of weight changes and output error. The proposed method was evaluated on 720 bladder cystoscopy images in four classes of blood in urine, benign and malignant masses. Based on the simulated results, the second proposed method (CNNs) achieved at least 17% decrease in error and increased the convergence speed of the proposed method in the classi cation of cystoscopy images, compared to the other competing methods.

Diabetic Retinopathy (DR) is one of the major complications of Diabetes, which is the injury to the retina of the diabetic patient and causes blindness if not diagnosed in early stages. Various machine learning classification and clustering approaches have been studied in literature with the purpose of improving the accuracy of the screening methods. Some of machine learning classification and clustering approaches are based on manually feature extraction of fundus images by image processing experts. In recent years, a new approach for image classification and diagnosis without using any manual feature extraction is proposed based on convolutional neural network (CNN). In medical imaging and diagnosis, training a deep CNN from scratch is difficult because it requires a large amount of labeled training data and the training procedure is a time consuming task to ensure proper convergence. Therefore, a very common method to train CNNs for medical diagnosis is fine-tuning a pre-trained CNN. In this paper, the pre-trained GoogleNet as a powerful CNN is employed on the Kaggle database for DR diagnosis from retinal images. To assess the efficacy of the clinical results, the proposed CNN algorithm is performed to diagnose DR from the images that are gathered from the the Navid-Didegan ophthalmology clinic.

Between different sources of renewable energy, wind energy, as an economical source of electrical power, has undergone a pronounced thriving. However, wind turbines are exposed to catastrophic failures, which may bring about irrecoverable ramifications. Therefore, they necessarily need condition monitoring and fault detection systems. These systems aim to reduce the number of attempts operators are required to do through the use of smart software algorithms, which are able to understand and decide with no human involvement. The gearboxes are usually responsible for the WT breakdowns. In this paper, convolutional neural networks are employed to develop an intelligent data-based condition-monitoring algorithm to differentiate healthy and damaged conditions that are evaluated with the national renewable energy laboratory (NREL) GRC database on the WT gearbox. Since it is much easier for convolutional neural networks to extract clues from high dimensional data, time-domain signals are embodied as texture images. Results show that the proposed methodology by utilizing a 2-D convolutional neural network for binary classification is capable of classifying the NREL GRC database with 99.76% accuracy.