Number Of Test Sets Research Articles

Comparing neural network architectures for non-intrusive speech quality prediction

Non-intrusive speech quality predictors evaluate speech quality without the use of a reference signal, making them useful in many practical applications. Recently, neural networks have shown the best performance for this task. Two such models in the literature are the convolutional neural network based DNSMOS and the bi-directional long short-term memory based Quality-Net, which were originally trained to predict subjective targets and intrusive PESQ scores, respectively. In this paper, these two architectures are trained on a single dataset, and used to predict the intrusive ViSQOL score. The evaluation is done on a number of test sets with a variety of mismatch conditions, including unseen speech and noise corpora, and common voice over IP distortions. The experiments show that the models achieve similar predictive ability on the training distribution, and overall good generalization to new noise and speech corpora. Unseen distortions are identified as an area where both models generalize poorly, especially DNSMOS. Our results also suggest that a pervasiveness of ambient noise in the training set can cause problems when generalizing to certain types of noise. Finally, we detail how the ViSQOL score can have undesirable dependencies on the reference pressure level and the voice activity level.

Research on Subsidence Prediction Method of Water-Conducting Fracture Zone of Overlying Strata in Coal Mine Based on Grey Theory Model

The development height and settlement prediction of water-conducting fracture zones caused by coal seam mining play an important role in the stability of overburden aquifers and the safety of roadways. Based on the engineering geological data of the J60 borehole in the Daliuta Coal Mine and the mining conditions of the 2−2 coal seam, China, this study established a similar material test model of mining overburden. The deformation characteristics of overlying strata in the mining process of coal seam were studied by using distributed optical fiber sensing technology, and the development height of water flowing fractured zone was determined. According to the equidistant sampling characteristics of Brillouin optical time domain reflection technology and the principle of the grey theory model, the settlement prediction model of the water-conducting fracture zone was established. By analyzing and comparing the prediction accuracy of the GM (1, 1) model, grey progressive model, and metabolic model, the optimal method for settlement prediction of the water-conducting fracture zone was discussed. The results show that, for the metabolic model, with the increase in the number of test sets and the decrease in the number of prediction sets, the mean square error ratio c and the small error probability p of the prediction accuracy evaluation parameters display a downward trend. The accuracy is related to the sudden change in the settlement of the water-conducting fracture zone caused by the breaking of the key stratum of the overlying rock. The optimal time of test sets selected for the best settlement prediction model is 7~8, and that of prediction sets selected is 5~6. For the GM (1, 1) model and the grey progressive model, the prediction accuracy of mining overburden subsidence is grade 4, which is not suitable for settlement prediction of water-flowing fractured zones.

Autonomous Classification of Digital Painting Images Based on Wireless Network

With the advancement of digital technology, the precise classification and rapid search of digital painting is important to the creation of digital painting image digitization. In order to deeply study the construction of a self-service classification system based on wireless network, this paper uses the parameter sample insertion method, comparative analysis method, and deconstruction column method, analyzes the independent classification algorithm, and simplifies the algorithm. And finally, we realize and create an efficient system that can independently classify digital images. When studying the best points of the classification efficiency of classification systems, the number of training sets is 2000 and the number of test sets is 120. The input data is classified as 3 layers, and the learning rate of each layer is 1.5, 0.06, and the sparsity is 0.04. The results show that when the sparsity value of both layers of CRBM is taken as 0.02, the classification results reach the best, indicating that here is the best result of this experiment. Further studying the classification system stability, the image from the F1 training set is labeled as 0 and the network data is labeled as 1, thus providing a training two-way model, and the ratio between the training and test sets is 3 : 1. The classification accuracy is quite stable. With the addition of more training data, the classification accuracy remains at around 50%. This shows that the system stability is guaranteed. Based on wireless network technology, a complete system is designed to classify digital images.

A method for pulsar searching: combining a two-dimensional autocorrelation profile map and a deep convolutional neural network

In pulsar astronomy, detecting effective pulsar signals among numerous pulsar candidates is an important research topic. Starting from space X-ray pulsar signals, the two-dimensional autocorrelation profile map (2D-APM) feature modelling method, which utilizes epoch folding of the autocorrelation function of X-ray signals and expands the time-domain information of the periodic axis, is proposed. A uniform setting criterion regarding the time resolution of the periodic axis addresses pulsar signals without any prior information. Compared with the traditional profile, the model has a strong anti-noise ability, a greater abundance of information and consistent characteristics. The new feature is simulated with double Gaussian components, and the characteristic distribution of the model is revealed to be closely related to the distance between the double peaks of the profile. Next, a deep convolutional neural network (DCNN) is built, named Inception-ResNet. According to the order of the peak separation and number of arriving photons, 30 data sets based on the Poisson process are simulated to construct the training set, and the observation data of PSRs B0531+21, B0540-69 and B1509-58 from the Rossi X-ray Timing Explorer (RXTE) are selected to generate the test set. The number of training sets and the number of test sets are 30 000 and 5400, respectively. After achieving convergence stability, more than 99% of the pulsar signals are recognized, and more than 99% of the interference is successfully rejected, which verifies the high degree of agreement between the network and the feature model and the high potential of the proposed method in searching for pulsars.

A benchmark study of ab initio gene prediction methods in diverse eukaryotic organisms

BackgroundThe draft genome assemblies produced by new sequencing technologies present important challenges for automatic gene prediction pipelines, leading to less accurate gene models. New benchmark methods are needed to evaluate the accuracy of gene prediction methods in the face of incomplete genome assemblies, low genome coverage and quality, complex gene structures, or a lack of suitable sequences for evidence-based annotations.ResultsWe describe the construction of a new benchmark, called G3PO (benchmark for Gene and Protein Prediction PrOgrams), designed to represent many of the typical challenges faced by current genome annotation projects. The benchmark is based on a carefully validated and curated set of real eukaryotic genes from 147 phylogenetically disperse organisms, and a number of test sets are defined to evaluate the effects of different features, including genome sequence quality, gene structure complexity, protein length, etc. We used the benchmark to perform an independent comparative analysis of the most widely used ab initio gene prediction programs and identified the main strengths and weaknesses of the programs. More importantly, we highlight a number of features that could be exploited in order to improve the accuracy of current prediction tools.ConclusionsThe experiments showed that ab initio gene structure prediction is a very challenging task, which should be further investigated. We believe that the baseline results associated with the complex gene test sets in G3PO provide useful guidelines for future studies.

Cross-covariance-based features for speech classification in film audio

As multimedia becomes the dominant form of entertainment through an ever increasing range of digital formats, there has been a growing interest in obtaining information from entertainment media. Speech is one of the core resources in multimedia, providing a foundation for the extraction of semantic information. Thus, detecting speech is a critical first step for speech-based information retrieval systems. This work focuses on speech detection in one of the dominant forms of entertainment media: feature films. A novel approach for voice activity detection (VAD) in film audio is proposed. The approach uses correlation to analyze associations of Mel Frequency Cepstral Coefficient (MFCC) pairs in speech and non-speech data. This information then drives feature selection for the creation of MFCC cross-covariance feature vectors (MFCC-CCs) which are used to train a random forest classifier to solve a binary speech/non-speech classification problem on audio data from entertainment media. The classifier performance is evaluated on a number of test sets and achieves a classification accuracy of up to 94%. The approach is also compared with state of the art and contemporary VAD algorithms, and demonstrates competitive results.

Feature Matching with Bounded Distortion

We consider the problem of finding a geometrically consistent set of point matches between two images. We assume that local descriptors have provided a set of candidate matches, which may include many outliers. We then seek the largest subset of these correspondences that can be aligned perfectly using a nonrigid deformation that exerts a bounded distortion. We formulate this as a constrained optimization problem and solve it using a constrained, iterative reweighted least-squares algorithm. In each iteration of this algorithm we solve a convex quadratic program obtaining a globally optimal match over a subset of the bounded distortion transformations. We further prove that a sequence of such iterations converges monotonically to a critical point of our objective function. We show experimentally that this algorithm produces excellent results on a number of test sets, in comparison to several state-of-the-art approaches.

A Local-stem-search Algorithm to Predict The RNA Secondary Structure*

RNA secondary structure predicting is a classical problem in bioinformatics and the optimal algorithms based on minimal free energy (MFE) criterion are the widely used methods. However, pseudoknots render the problem of computing the RNA MFE structure with pseudoknot becomes a NP-hard problem. A heuristic algorithm——StemFind to predict RNA secondary structure with pseudoknot was presented. The algorithm regard stem as the basic search unit, adopting heuristic search strategy, and search the most possible RNA secondary structure in stem combination space. The StemFind algorithm to a large number of test sets was applied. Performance evaluation demonstrates that StemFind not only outperforms the well-known optimal and heuristic algorithms in overall sensitivity and specificity but also requires significantly less time than the optimal algorithm.

Temperature tests on package lid seal materials over one year duration

The design of lid seal system of any package transporting irradiated nuclear fuel is a major aspect of the containment safety case under normal and accident conditions. Consequently, when BNFL, now known as British Nuclear Group Sellafield, BN-GS, decided on a change in lid seal material, year long proving trials on the new material were conducted, these simulating actual service and accident temperature conditions. Several years ago, BN-GS with the Rubber and Plastic Research Association (RAPRA) developed an elastomer called EPDM 30H which could perform at temperatures below −40°C and had excellent resistance to radiation, as later proven by long term testing. EPDM 30H material has been used in many site applications, with its first transport package application being on the Excellox 6 type, commissioned in 1991. Towards the end of the 1990s, BN-GS became aware that the fluorocarbon grade of lid seal material, used on several other packages, may be discontinued by the manufacturers and so they took the decision to replace the seals on the NTL 11 type packages with EPDM 30H. Consequently, BN-GS embarked on an extensive programme of seal testing which went beyond any it had previously carried out. Comprehensive data were available on radiation resistance and performance at low temperatures, but additional data were needed on its behaviour at elevated temperature over periods of about one year. A number of test sets were assembled comprising seals in representative lid seal grooves. Most of the test sets were fitted with EPDM 30H seals but another grade of EPDM seal was also included. Two test sets were continuously maintained at temperatures of 120 and 150°C, for one year. Other test sets were maintained at 60, 90, 120, 140, 160 and 180°C, but periodic inspection and compression set measurements were taken. At the end of one year the continuously heated seals were subjected to a thermal cycle corresponding to the thermal accident safety cases studies. These tests demonstrated that the EPDM 30H material had lower compression set characteristics than the other grade of EPDM and that EPDM 30H was a suitable seal material for all irradiated LWR fuel transport packages currently operated BN-GS.

A Compressed‐Air Undercarriage

A LICENCE has lately been obtained to manufacture in England the “Faudi” compressed‐air landing legs. Of German origin, this type of undercarriage strut has already been widely used in that country, where it is fitted on the Luft Hansa Junkers used on the Croydon service and the Heinkel H.E.70. It is also well known in France and is being manufactured in Japan. In England it is being made by Turner's Motor Manufacturing Co., Ltd., of Wulfruna Works, Lever Street, Wolverhampton, who are the concessionnaires. This firm is already in production and a number of test sets are being supplied to various aircraft firms.