Search

Automatic speech recognition is a research field for enabling computers to receive voice feedback from humans and to translate it with the greatest probability of correctness. Speech is the easiest way to communicate the message in basic context. Speech recognition is referring the allowing of machines to perceive and accept speech information from humans for the greatest probability of accuracy. Speech recognition will play an ever more important role in the future of human with machine interfaces. Arabic speech recognition (ASR) is a most complex job and has received little attention than other languages. In this paper we reviewing the most important deep learning methods used in speech recognition in Arabic language. Through the search the main finding in this review that the Arabic is one of the most spoken languages and least highlighted in terms of speech recognition. One of the emerging techniques is using neural networks with deep learning for speech recognition. Deep Neural Networks (DNNs) have reached to suitable performance, Recurrent Neural Networks (RNN) is one of the best models applied for sequential data and convolution neural network (CNN) has obtained a best accuracy rate in speech recognition. [ABSTRACT FROM AUTHOR]

In recent times automated lip reading has gained more research attention. Much advancement are being developed in the area using various deep learning algorithms. Automated lip reading can be performed with or without audio. When lip movements are identified without sound of speech it is often referred as visual speech recognition. One of the drawbacks in visual speech recognition is the detecting words that have similar lip movement. Visemes are often referred as speech sounds that look same for different words that have similar lip movement. In this paper an Inception encoder decoder network is proposed for visual lip-reading. The model is developed to detect the lip-read sentences from a variety of vocabulary and other sentences that are not included in model training. In proposed method visemes are classified and used as a classification method for detecting lip reading sentences. Detected visemes are converted to sentences using perplexity analysis. Proposed method is lexicon free and totally based on visual cues. Visemes are used. The model has been experimented on Lip Reading Sentences 3 TED benchmark dataset containing various challenging videos from TED and TEDX talks. Besides the LRS-3 dataset additionally various experiments have been performed with videos of varying illumination. Proposed model achieved state of art results compared with current lip-reading models. Experimental results show that developed model performs significantly with 13% lower error rate with more robustness to varying illumination. [ABSTRACT FROM AUTHOR]

Visually impaired person are facing lot of challenges to prefer online shopping. These persons are more likely to experience restrictions like mobility, less educational achievement and independence, Moreover increased risk of falling down, injuries, fractures, cognitive deficits, poor mental health, and social isolation. This paper pave a way to overcome this situation an enhanced system is being developed for online shopping. It provides simplest solutions with various options for a person whether he/she is visually impaired or not. The developed system is similar to shopping cart like amazon and flipkart in. it possesses proper authentication of a user to perform various activities. What makes it different is, it enables the facility of online shopping for visually impaired person through speech recognition. This is done using deep learning and machine learning algorithm. Here the input will be user's voice and this voice will be automatically decoded into text by using the algorithm. [ABSTRACT FROM AUTHOR]

In recent times, deep learning algorithms have achieved highly competitive performance in emotion recognition tasks and speech recognition tasks for representing complex relationships and extracting abstract features from speech files. In the proposed technique, we apply Deep belief networks with shallow network to speech files and used to train the machine with training data set to classify them in different categories of emotion states. First speech wave files are pre-processed then parallel training sample sets are established and which takes as input to Deep belief network for automatically extraction of speech emotion features. Finally, classification is performed by shallow neural network to recognize the final results. Accuracy of this system depend upon the amount of pre-training dataset used and number of hidden layers, epochs are implemented to automatically extract the features and for classification process. In this paper we compare three methodologies to identify emotions and experimental results shows that combination of deep belief network (DBN) and shallow network can give significant improvement in accuracy of emotion recognition system. [ABSTRACT FROM AUTHOR]

Emotion speech recognition is a developing field in machine learning. The main purpose of this field is to produce a convenient system that is able to effortlessly communicate and interact with humans. The reliability of the current speech emotion recognition systems is far from being achieved. However, this is a challenging task due to the gap between acoustic features and human emotions, which rely strongly on the discriminative acoustic features extracted for a given recognition task. The speech signals were process with information which is divided into two main categories, linguistic and paralinguistic; emotions belong to the latter tree. The aim of this work is to develop a system that can understand paralinguistic information for paramount better human-machine interactions. A different extracted features like MFCC as well as feature classifications methods like HMM, GMM, LTSTM and ANN were used. In this paper, an improved architecture of CNN for speech emotion recognition were implemented. The main finding that the proposed CNN model achieved 93.96% accuracy rate in detecting emotions. [ABSTRACT FROM AUTHOR]

Recent high-profile successes in machine learning have found solutions to problems that were long-thought to be decades away and has generated renewed interest in artificial intelligence (AT) and machine learning (ML) research. For example, the recent success and rapid commercialization of deep learning has catapulted technical achievements in many fields, including computer vision, speech recognition, games, and machine translation. These successes present new opportunities for advancing nondestructive evaluation (NDE) techniques. This paper reviews the fundamental concepts of ML and their underlying connections to statistics. We then discuss past applications and methods of ML for NDE. We also describe the still unresolved challenges for this field, such as the unavailability of reliable training data. We then discuss current research in ML for NDE that is directed toward solving these challenges. Finally, we outline how the most recent advances in ML, such as deep learning and transfer learning, have the potential to revolutionize how we design future NDE solutions. [ABSTRACT FROM AUTHOR]

Deep learning is an emerging technology that is one of the most promising areas of artificial intelligence. Great strides have been made in recent years which resulted in increased efficiency with regards to many applications, including speech. Despite this, an environmentally Robust Speech Recognition system is still far from being achieved. In this article, an investigation of previous work has been conducted. The use of deep learning in speech recognition was analyzed and a proper deep learning architecture was identified. A method using convolutional neural network (CNN) is used with the aim of enhancing the performance of speech recognition systems (SRS). This study found that this CNN-based approach achieves a 94.32% validated accuracy [ABSTRACT FROM AUTHOR]