Emotional voice conversion using deep neural networks with MCC and F0 features

Voice conversion (VC) is a task that alters the voice of a person to suit different styles while conserving the linguistic content. Previous state-of-the-art technology used in VC was based on the sequence-to-sequence (seq2seq) model, which could lose linguistic information. There was an attempt to overcome this problem using textual supervision; however, this required explicit alignment, and therefore the benefit of using seq2seq model was lost. In this study, a voice converter that utilizes multitask learning with text-to-speech (TTS) is presented. By using multitask learning, VC is expected to capture linguistic information and preserve the training stability. This method does not require explicit alignment for capturing abundant text information. Experiments on VC were performed on a male-Korean-emotional-text-speech dataset to convert the neutral voice to emotional voice. It was shown that multitask learning helps to preserve the linguistic content.

Algorithms based on deep learning are taking off as solutions to several problems in numerous fields. Numerous deep learning architectures like convolutional neural networks, recurrent neural networks, deep belief networks, artificial neural networks and deep neural networks have been used in fields such as audio recognition, machine vision, processing of natural language, recognition of speech, social network filtering, computer vision, translating languages and bioinformatics. One such application of deep learning architectures is object detection. Object detection in satellite and aerial imagery undertakes an important role in a long list of applications and has allured a lot of attention which resulted in exceptional progress in recent years. It has abundant uses in various fields such as surveillance, military, national security, urban planning, environmental applications, traffic control and forestry. Several issues like varying size, background and orientation of the target object make the automatic detection of target in satellite imagery a challenging issue. Recent research in this field has given better speed and accuracy to the detection of the target object. With the rise in the data being generated by the Earth Observation programs, it is becoming much more imperative to find better and efficient ways to exploit this data by improving upon the existing Deep Learning research. In this paper, we bring forth a wide-ranging analysis of the recent advancements in methods of object detection based on deep learning to recognize the objects in satellite imagery.KeywordsDeep learningObject detectionSatellite imageryRemote sensingYou Only Look Once (YOLO)You Only Look Twice (YOLT)Single Shot Detector (SSD)Convolutional Neural Networks (CNN)Faster Region-based Convolutional Neural Networks (Faster RCNN)Hybrid Deep Convolutional Neural Network (HDNN)Satellite Imagery Multiscale Rapid Detection with Windowed Networks (SIMRDWN)

Recently, there have been significant advances in deep neural networks (DNNs) and they have shown distinctive performance in speech recognition, natural language processing, and image recognition. In this paper, we explore DNNs to push the limit for branch prediction. We treat branch prediction as a classification problem and employ both deep convolutional neural networks (CNNs), ranging from LeNet to ResNet-50, and deep belief network (DBN) for branch prediction. We compare the effectiveness of DNNs with the state-of-the-art branch predictors, including the perceptron, our prior work, Multi-poTAGE+SC, and MTAGE+SC branch predictors. The last two are the most recent winners of championship branch prediction (CBP) contests. Several interesting observations emerged from our study. First, for branch prediction, the DNNs outperform the perceptron model as high as 60-80%. Second, we analyze the impact of the depth of CNNs (i.e., number of convolutional layers and pooling layers) on the misprediction rates. The results confirm that deeper CNN structures can lead to lower misprediction rates. Third, the DBN could outperform our prior work, but not outperform the state-of-the-art TAGE-like branch predictor; the ResNet-50 could not only outperform our prior work, but also the Multi-poTAGE+SC and MTAGE+SC.

Handwritten digit recognition is the intelligence of computers to recognize digits written by humans. But it becomes one of the most challenging tasks for machines as handwritten digits are not perfect and can be made with many different: flavors, size, thickness. Thus, as a solution to this problem, Handwriting digit recognition model comes into picture. Many machine learning techniques have been employed to solve the handwritten digit recognition problem. This paper focuses on Neural Network (NN) approaches. Among the three famous NN approaches: deep neural network (DNN), deep belief network (DBN) and convolutional neural network (CNN), the specialization of CNN as compared to other NN of being able to detect pattern is what makes it so useful for recognizing handwritten digits. Humans can very easily see, read & write any handwritten digits, when written in proper format. Even if the digits are not written in proper format we can use our logic and predict what digit it could be. But It is a hard task for the machine to recognize handwritten digits as these are not perfect and can be made with many different flavors. Thus handwritten digit recognition is the solution to this problem which uses the image of a digit and recognizes the digit present in the image. The handwritten digit recognition is the ability of computers to recognize human handwritten digits. The goal is to build a model that can efficiently and reliably recognize the digits and output the proper result. Amongst all the other neural networks, working and implementing a model using Convolution Neural Network gives out the most precise results. It is most popularly used for analyzing images as well as for other data analysis or classification problems. CNN has hidden layers called convolutional layers. These layers work the same way as other layers do but here we need to specify the no of filters each layer should have These filters are actually what detects the pattern. Patterns could be edges, corners, circles or any complex other objects like eyes, ears or even deeper full dogs, cats, etc. Thus, the specialization of CNN as compared to other NN of being able to detect patterns is what makes it so useful for recognizing handwritten digits Key Word: Neural Networks; Convolutional Neural Networks (CNN); Image Processing; Optical Character Recognition (OCR); MNIST Dataset

Deep learning has been successfully applied to the field of fault diagnosis in recent years. Due to the advantages of deep belief network (DBN) in fitting nonlinear complex systems and the ability of wavelet analysis in time-frequency analysis, in this paper, an improved fault diagnosis method based on a deep wavelet neural network (DWNN), which combines the DBN with morlet activation functions, is proposed for fault diagnosis of reciprocating compressor. A five-layer DBN using sigmoid, tanh, rectified linear unit (ReLU) and morlet wavelet functions as the activation functions of hidden layers separately is proposed for fault diagnosis of reciprocating compressor. As the contrast, a three-layer back propagation neural network (BPNN) using the same four activation functions separately is proposed for fault diagnosis of reciprocating compressor. The experimental results show that, the fault diagnosis rate of five-layer DBN is higher than the three-layer BPNN. The method based on DWNN can make the fault diagnosis rate reach 100% within short time. Compared with using other activation functions, the DWNN architecture requires less epochs to train the model.

The paper is focused on the idea to demonstrate the advantages of deep learning approaches over ordinary shallow neural network on their comparative applications to image classifying from such popular benchmark databases as FERET and MNIST. An autoassociative neural network is used as a standalone program realized the nonlinear principal component analysis for prior extracting the most informative features of input data for neural networks to be compared further as classifiers. A special study of the optimal choice of activation function and the normalization transformation of input data allows to improve efficiency of the autoassociative program. One more study devoted to denoising properties of this program demonstrates its high efficiency even on noisy data. Three types of neural networks are compared: feed-forward neural net with one hidden layer, deep network with several hidden layers and deep belief network with several pretraining layers realized restricted Boltzmann machine. The number of hidden layer and the number of hidden neurons in them were chosen by cross-validation procedure to keep balance between number of layers and hidden neurons and classification efficiency. Results of our comparative study demonstrate the undoubted advantage of deep networks, as well as denoising power of autoencoders. In our work we use both multiprocessor graphic card and cloud services to speed up our calculations. The paper is oriented to specialists in concrete fields of scientific or experimental applications, who have already some knowledge about artificial neural networks, probability theory and numerical methods.

PurposeUp to date development in sentiment analysis has resulted in a symbolic growth in the volume of study, especially on more subjective text types, namely, product or movie reviews. The key difference between these texts with news articles is that their target is defined and unique across the text. Hence, the reviews on newspaper articles can deal with three subtasks: correctly spotting the target, splitting the good and bad content from the reviews on the concerned target and evaluating different opinions provided in a detailed manner. On defining these tasks, this paper aims to implement a new sentiment analysis model for article reviews from the newspaper.Design/methodology/approachHere, tweets from various newspaper articles are taken and the sentiment analysis process is done with pre-processing, semantic word extraction, feature extraction and classification. Initially, the pre-processing phase is performed, in which different steps such as stop word removal, stemming, blank space removal are carried out and it results in producing the keywords that speak about positive, negative or neutral. Further, semantic words (similar) are extracted from the available dictionary by matching the keywords. Next, the feature extraction is done for the extracted keywords and semantic words using holoentropy to attain information statistics, which results in the attainment of maximum related information. Here, two categories of holoentropy features are extracted: joint holoentropy and cross holoentropy. These extracted features of entire keywords are finally subjected to a hybrid classifier, which merges the beneficial concepts of neural network (NN), and deep belief network (DBN). For improving the performance of sentiment classification, modification is done by inducing the idea of a modified rider optimization algorithm (ROA), so-called new steering updated ROA (NSU-ROA) into NN and DBN for weight update. Hence, the average of both improved classifiers will provide the classified sentiment as positive, negative or neutral from the reviews of newspaper articles effectively.FindingsThree data sets were considered for experimentation. The results have shown that the developed NSU-ROA + DBN + NN attained high accuracy, which was 2.6% superior to particle swarm optimization, 3% superior to FireFly, 3.8% superior to grey wolf optimization, 5.5% superior to whale optimization algorithm and 3.2% superior to ROA-based DBN + NN from data set 1. The classification analysis has shown that the accuracy of the proposed NSU − DBN + NN was 3.4% enhanced than DBN + NN, 25% enhanced than DBN and 28.5% enhanced than NN and 32.3% enhanced than support vector machine from data set 2. Thus, the effective performance of the proposed NSU − ROA + DBN + NN on sentiment analysis of newspaper articles has been proved.Originality/valueThis paper adopts the latest optimization algorithm called the NSU-ROA to effectively recognize the sentiments of the newspapers with NN and DBN. This is the first work that uses NSU-ROA-based optimization for accurate identification of sentiments from newspaper articles.

We study the eigenvalue distribution of random matrices pertinent to the analysis of deep neural networks. The matrices resemble the product of the sample covariance matrices, however, an important difference is that the analog of the population covariance matrix is now a function of random data matrices (synaptic weight matrices in the deep neural network terminology). The problem has been treated in recent work [J. Pennington, S. Schoenholz and S. Ganguli, The emergence of spectral universality in deep networks, Proc. Mach. Learn. Res. 84 (2018) 1924–1932, arXiv:1802.09979] by using the techniques of free probability theory. Since, however, free probability theory deals with population covariance matrices which are independent of the data matrices, its applicability in this case has to be justified. The justification has been given in [L. Pastur, On random matrices arising in deep neural networks: Gaussian case, Pure Appl. Funct. Anal. (2020), in press, arXiv:2001.06188] for Gaussian data matrices with independent entries, a standard analytical model of free probability, by using a version of the techniques of random matrix theory. In this paper, we use another version of the techniques to extend the results of [L. Pastur, On random matrices arising in deep neural networks: Gaussian case, Pure Appl. Funct. Anal. (2020), in press, arXiv:2001.06188] to the case where the entries of the data matrices are just independent identically distributed random variables with zero mean and finite fourth moment. This, in particular, justifies the mean field approximation in the infinite width limit for the deep untrained neural networks and the property of the macroscopic universality of random matrix theory in this case.

Experimental and DBN-Based neural network extraction of radiation attenuation coefficient of dry mixture shotcrete produced using different additives

Automatic speaker verification (ASV) faces domain shift caused by the mismatch of intrinsic and extrinsic factors, such as recording device and speaking style, in real-world applications, which leads to severe performance degradation. Since single-speaker multi-condition (SSMC) data is difficult to collect in practice, existing domain adaptation methods are hard to ensure the feature consistency of the same class but different domains. To this end, we propose a cross-domain data generation method to obtain a domain-invariant ASV system. Inspired by voice conversion (VC) task, a StarGAN based generative model first learns cross-domain mappings from SSMC data, and then generates missing domain data for all speakers, thus increasing the intra-class diversity of the training set. Considering the difference between ASV and VC task, we renovate the corresponding training objectives and network structure to make the adaptation task-specific. Evaluations on achieve a relative performance improvement of about 5-8% over the baseline in terms of minDCF and EER, outperforming the CNSRC winner’s system of the equivalent scale.

Through the looking glass: Seeing events in power systems dynamics

Deep structured learning or hierarchical learning or deep learning in short is part of the family of machine learning methods which are themselves a subset of the broader field of Artificial Intelligence. Each successive layer uses the output from the previous layer as input. Deep neural networks, deep belief networks and recurrent neural networks have been applied to fields such as computer vision, speech recognition, natural language processing, audio recognition, social network filtering, machine translation, and bioinformatics where they produced results comparable to and in some cases better than human experts have. CNN are applied for deep feature extraction and LSTM is used to detect the class based on extracted features. Deep learning technique that would sort the fruit into abnormal and normal based on the feature such as fruit color, number of fruit spots , and shapes of the fruit, The proposed system achieved an accuracy of 98.17%, The susceptibility of disease detection was less with lesser availability of enhanced detection methods for detecting disease in earlier stages. The issue with various existing algorithms is that the accuracy was reduced so some problems not eliminated. Deep learning delivers methodologies, approaches and functionalities that can help to resolve analytic and predictive analysis accurately.

Deep learning (also known as deep structured learning, hierarchical learning, or deep machine learning) is a branch of machine learning based on a set of algorithms that attempt to model high-level abstractions in data by using a deep graph with multiple processing layers, composed of multiple linear and nonlinear transformations. Deep learning has been characterized as a class of machine learning algorithms with the following characteristics [257]: • They use a cascade of many layers of nonlinear processing units for feature extraction and transformation. Each successive layer uses the output from the previous layer as input. The algorithms may be supervised or unsupervised and applications include pattern analysis (unsupervised) and classification (supervised). • They are based on the (unsupervised) learning of multiple levels of features or representations of the data. Higher-level features are derived from lower-level features to form a hierarchical representation. • They are part of the broader machine learning field of learning representations of data. • They learn multiple levels of representations that correspond to different levels of abstraction; the levels form a hierarchy of concepts. These definitions have in common: multiple layers of nonlinear processing units and the supervised or unsupervised learning of feature representations in each layer, with the layers forming a hierarchy from low-level to high-level features. Various deep learning architectures such as deep neural networks, convolutional deep neural networks, deep belief networks (DBN), and recurrent neural networks have been applied to fields like computer vision, automatic speech recognition, natural language processing, audio recognition, and bioinformatics where they have been shown to produce state-of-the-art results on various tasks. In this chapter we start in Section 7.1 with an introduction, giving a brief history of this field, the relevant literature, and its applications. Then we study some basic concepts of deep learning such as convolutional neural networks, recurrent neural networks, backpropagation algorithm, restricted Boltzmann machines, and deep learning networks in Section 7.2. Then we illustrate three examples for Apache Spark implementation for mobile big data (MBD), user moving pattern extraction, and combination with nonparametric Bayesian learning, respectively, in Sections 7.3 through 7.5. Finally, we have summary in Section 7.6.

A novel approach to pitch transposition is presented, which allows to preserve the spectral envelope or to modify it in a voice conversion task. Pitch-shifted output is synthesized directly from the input signal using the phase vocoder, without explicitly estimating the fundamental frequency, which is often prone to errors. Frequency band extension technique is proposed to address the problem of bandwidth reduction when scaling pitch down. Artificial neural network was applied to transform the signal's spectral envelope in the full voice conversion task. Listening tests showed that 55% of listeners preferred the quality of the pitch modification using the phase vocoder over the one offered by a parametric approach.

This paper proposes a new voice conversion (VC) task from human speech to dog-like speech while preserving linguistic information as an example of human to non-human creature voice conversion (H2NH-VC) tasks. Although most VC studies deal with human to human VC, H2NH-VC aims to convert human speech into non-human creature-like speech. Non-parallel VC allows us to develop H2NH-VC, because we cannot collect a parallel dataset that non-human creatures speak human language. In this study, we propose to use dogs as an example of a non-human creature target domain and define the "speak like a dog" task. To clarify the possibilities and characteristics of the "speak like a dog" task, we conducted a comparative experiment using existing representative non-parallel VC methods in acoustic features (Mel-cepstral coefficients and Mel-spectrograms), network architectures (five different kernel-size settings), and training criteria (variational autoencoder (VAE)- based and generative adversarial network-based). Finally, the converted voices were evaluated using mean opinion scores: dog-likeness, sound quality and intelligibility, and character error rate (CER). The experiment showed that the employment of the Mel-spectrogram improved the dog-likeness of the converted speech, while it is challenging to preserve linguistic information. Challenges and limitations of the current VC methods for H2NH-VC are highlighted.