Search results

With extensive experiments on one representative SD dataset of three speakers' recordings, our method achieves better accuracy than DTW, where 0.5% relative reduction of error rate (RRER) on clean speech data and 7.5% RRER on noisy and bad recording speech data.
In MWDTW, we define a similar index for each merged training data as CIDTW, and then perform DTW alignment on merged speech data with index identified.
Compared with original DTW, our new weighted method achieves better accuracy both on clean data with 0.5% relative reduction of error rate (RRER) and on noisy data with7.5% RRER.
Results Data Preparation.
Given ten time recordings for each name, one audio file of a certain name is randomly picked out as training data, the other nine files are testing data.

At each vertical line, water quality parameters such as dissolved oxygen (DO), salinity, turbidity, pH value, electrical conductivity and oxidation reduction potential (ORP) were realtimely measured with YSI 6600 V2-4-M multi-parameter water quality sonde.
Variation trend of turbidity at surface and bed layers in flood plain is identical, and the point data are almost same.
We can easily see from the Figure that there is no distinct discrepancy in the conductivity in main channel each cross-section streamwise, all point data overlap one another.
Fig.4 Variation in electrical conductivity Oxidation Reduction Potential Fig.5 shows that streamwise variation of the ORP at surface, intermediate and bed layers in main channel.
Fig.5 Variation in oxidation reduction potential Salinity of Water Body As mentioned above, Fuchunjiang and Hangzhou reaches refer to tidal river.

According to the analysis model, in case of keeping same energy efficiency of device, the higher technological level increases, the easier the R increases; even if keep the same melting and heat treatment devices, significant reduction of production energy consumption would be implemented just depending on the production yield increase.
Russian scholar Fromme claimed that by restructuring the production can lead to a 30% reduction in energy demand through case study [5].
Based on the melting devices nowadays in China is basically cupola in cast irons production, according to the data from many plants and enterprises, the average energy consumption volumes for melting and heat treatment are 50% and 30% of total consumption respectively.
Figure 2 Relationship between yield increase and unit energy consumption reduction The current average process yield in Chinese cast irons production is about 65~80%, reject rate is 9~14%, and the corresponding data abroad are 80%~90% and less than 5% respectively.
Therefore, even if keep the same melting and heat treatment devices, significant reduction of production energy consumption would be implemented just depending on the production yield increase.

This test has shown the adequacy of the model to real data.
Training of a neural network was provided on the basis of a database which included daily data of furnace operation parameters.
By the repeating training with a variety of actual data, sequential approach to the optimal values of the weights took place.
The adequacy of the model was tested using data which were not included in the training set.
Actions of considered factors on the coefficient of burden resistance are characterized by the data given in the Table 1.

Due to the lack of available data on domestic water distribution systems, the macroscopic model is used to model the water distribution system in this study.
Improved PSO based on AHP-RS and RBF Neural Network AHP-RS and RBF neural networks are loosely based on the neural structure of the brain which provides the ability to learn from the input data they are given and then apply this to unknown data, in effect they can generalize and associate unknown data.
RS is used to Discreting Sample Data.
Then, the data of sample 8 is used to be monitoring sample.
As mentioned previously, 48 data groups are kept for testing.

The analyzed OFE copper has been industrially engineered, being successively hot rolled and cold rolled to ∆=70 % and ∆=90 % reduction.
The software OIMTM analysis from the TSL Company has been used for the qualitative study of the EBSD data.
Intergranular nucleation and growth have been observed in the beginning of the recrystallization process for both reduction levels (70 and 90 %).
a),b) 70 % reduction. c),d) 90 % reduction.
For this reduction level, the second mechanism observed corresponds to the nucleation and growth of strongly recovered cube subgrains.

The essence of this research work is to determine individual stiffness of 3D space frame members using reduction member method.
By applying reduction member method, the individual stiffness of all member can be determined.
Fig. 2 : Flow process to find individual stiffness of 3D space frame joint members 2.1 Experimental Data An experiment for 3D space frame structural joint is carried out to find the relationship between force and displacement.
By applying reduction member method, the individual stiffness of all members can be calculated.
Figure 6: Load versus deformation for experimental and simulation Fig. 7: Equivalent stiffness of all members 4.0 Conclusion From this research work, reduction member method is proposed to predict individual stiffness for 3D space frame thin walled structural joint.

From the experimental ellipsometric data (see Fig. 3) were determined the thicknesses of samples consisting of 1, 3 and 5 layers (11 nm, 36 nm, 60 nm).
The conductivity of GO is affected by annealing (by the reduction of GO) for both polarities.
Analysis revealed that the experimental data describes the electrical properties of the bulk material and the contact of material with Al electrode (Schottky barrier) – i.e. the bulk conductivity (GB) and geometric capacity (Cg ≈ CB), resp. the conductivity (GS) and capacity (CS) of the interface.
Ruoff, Synthesis of graphene-based nanosheets via chemical reduction of exfoliated graphite, Carbon 45 (2007) 1558-1565 [4] S.
Cheng, The reduction of graphene oxide, Carbon 50 (2012) 3210-3228

The high performance of the Co10-Ni10/CeO2 was attributed to enhancement of surface Ce4+ reduction and fine particles of metal.
As CH4 showed signals in TCD and FID simultaneously, total analysis of the products was given and data with carbon balance over 95% were adopted.
The first reduction peak temperature at 320oC was attributed to the reduction of Co3+ to Co2+ and the second at 380oC was assigned to the reduction of Co2+ to Co0.
For ceria, two reduction peaks could be observed.
The first broad peak at 400oC was due to the reduction of surface and subsurface Ce4+ to Ce3+, while the second peak at over 800oC corresponded to the reduction of bulk Ce4+ to Ce3+.

However, its data transmission rate is limited to a few bits per second due to attenuations and spreading of pulses [3].
In this case, the data transmission along the drillstring via acoustic stress waves offers another communication possibility.
As early as 1948, acoustic telemetry was attempted to be applied to downhole sensor data communication along drillstring.
Summary Acoustic telemetry is a promising technique for acquiring downhole sensor data in real time.
Johnson, Limits on data communication along the drillstring using acoustic waves, SPE Reservoir Evaluation and Engineering. 11, 1(2008) 141-146