Generally, the performance of endpoint detection is affected by the noise. In this paper, we propose a novel two-layer decision model based on noise classification to detect the activity voice robustly. The training processing mainly contains two steps: firstly, we employ the noisex-92 database, which consists of different types of pure noise, to train a BP neural network in order to classify the noise type precisely, secondly, we train BP neural networks for each noise type covering large range of signal noise ratio (SNR). In the testing phase, we assume that the short period of silence at the beginning of the signal contains features for noise and utilize them to get the noise type. Then, we use the classifier corresponding to the noise type to detect activity voice. We conduct experiments on TIMIT corpus for 5 noise types under 7 SNR conditions. And experimental results show that our method outperforms global classifier, especially in low SNR condition.
Morphology,structure and phases of TiH2-47Al-0.2Si-10Nb and TiH2-47Al-0.2Si-12Nb powders in the process of mechanical alloying(MA) were investigated.TiAl-based alloys were prepared using the powders milled for 30 h by spark plasma sintering(SPS) and vacuum sintering.Influence of ball milling of the powders on microstructure of the TiAl-based alloys was studied by X-ray diffraction(XRD),scanning electron microscopy(SEM) and transmission electron microscopy(TEM).The results show that amorphous phase and TiAl intermetallic compounds formed during the high energy mechanical ball milling,and the particle size reached about 10 nm for the powders after milling for 30 h.The microstructure consisted of fine and homogeneous TiAl and Ti3 Al phases with grain size of 1 μm is obtained for the TiAl-based alloy prepared by SPS and vacuum sintering.
Powder characteristics have a significant influence on selective laser melting (SLM) process. In this research, particle size, surface contour and the microstructure of IN718 powder used in SLM were tested, and the physical and chemical changes during heating process of the powder were investigated as well. The shape characteristic and fluid properties of IN718 powder were defined and measured. It is shows that most particles of IN718 powder exhibited nearly spherical shapes, and the particle size is about 22.95µm. On this basis, IN718 sample was fabricated by SLM process using the powder and the microstructure of the fabricated sample was investigated. The results indicated that the sample has a dense microstructure, no macro-defects but a few micropores and porosity. Compared with the powder, there was no new phase formed in the sample and melting point of the sample was little less than that of the powder.
Abstract The architecture of sandy braided rivers, especially the mid-channel bars and the internal intercalations, plays a key role in the enhancement of remaining oil. Based on ancient outcrops, modern deposits and geological models, taken P oilfield as an example, this paper studies the different hierarchies of the internal architectural elements to describe the heterogeneity of sandy braided fluvial reservoir qualitatively and quantitatively, and proposes techniques to characterize the architecture of underground braided fluvial reservoirs. Based on the proposed distribution pattern and the related characterization techniques, the different controls of architecture elements on the distribution of remaining oil are analyzed: (1) sandy braided rivers have three filling patterns, dominated by sandy fillings. The mid-channel bars and braided channels are distributed in the pattern of wide bars and narrow channels, while the silting layers within mid-channel bars have dome-like and horizon-like distribution patterns; (2) the architectures are characterized from 5th to 3rd-order hierarchies successively under the principle of thickness control, scale constraint and multidimensional interaction. Single braided belts can be classified by three identification marks (the average width/thickness ratio is about 178, the width ratio of braided channels and mid-channel bars is about 1:3.8, and the width/length ratio of mid-channel bars is about 1:2.2); (3)Single-period silting layers are nearly horizontally distributed within mid-channel bars and multi-period ones are parallel with each other, generally 2-4 silting layers; (4)Reservoir architectural elements of different hierarchies exert different controls on the distribution of remaining oil. The intercalations control the remaining oil directly in the hierarchy of braided belts. In the hierarchy of mid-channel bars, semi-muddy filling channels have enriched in remaining oil due to relatively poor watered out degree as a result of disconnected sand bodies in the upper part with connected sand bodies in the lower part. Influenced by silting layers and petrophysical boundaries, the remaining oil within mid-channel bars is mainly distributed as “segment-like” enrichment at the top of vertical aggraded bodies under the silting layers and at the margin of poor water flooding. Overall, the research not only has good applications in the integral adjustment schemes of mature wells and development designs of news wells in oilfields, but also provides geological basis for well deployment of adjusting wells, especially for producing remaining oil in horizontal wells.
Microbial community samples have been accumulating at a speed faster than ever, with hundreds of thousands of samples been sequenced each year. Mining such a huge amount of multisource heterogeneous data is becoming an increasingly difficult challenge, so efficient and accurate compare and search of samples is in urgent need: faced with millions of samples in the data repository, traditional sample comparison and search approaches fall short in speed and accuracy.
This paper proposes a novel cache architecture suitable for merged DRAM/logic LSIs, which is called "dynamically variable line-size cache (D-VLS cache)". The D-VLS cache can optimize its line-size according to the characteristic of programs, and attempts to improve the performance by exploiting the high on-chip memory bandwidth. In our evaluation, it is observed that the performance improvement achieved by a direct-mapped D-VLS cache is about 27%, compared to a conventional direct-mapped cache with fixed 32-byte lines.
Abstract Microbiome samples are accumulating at an unprecedented speed. As a result, a massive amount of samples have become available for the mining of the intrinsic patterns among them. However, due to the lack of advanced computational tools, fast yet accurate comparisons and searches among thousands to millions of samples are still in urgent need. In this work, we proposed the Meta-Prism method for comparing and searching the microbial community structures amongst tens of thousands of samples. Meta-Prism is at least 10 times faster than contemporary methods serving the same purpose and can provide very accurate search results. The method is based on three computational techniques: dual-indexing approach for sample subgrouping, refined scoring function that could scrutinize the minute differences among samples, and parallel computation on CPU or GPU. The superiority of Meta-Prism on speed and accuracy for multiple sample searches is proven based on searching against ten thousand samples derived from both human and environments. Therefore, Meta-Prism could facilitate similarity search and in-depth understanding among massive number of heterogenous samples in the microbiome universe. The codes of Meta-Prism are available at: https://github.com/HUST-NingKang-Lab/metaPrism.
Download This Paper Open PDF in Browser Add Paper to My Library Share: Permalink Using these links will ensure access to this page indefinitely Copy URL Copy DOI
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)