Robust Speaker Recognition in Noisy Conditions

This paper investigates the problem of speaker identi-fication and verification in noisy conditions, assuming that speechsignals are corrupted by environmental noise, but knowledgeabout the noise characteristics is not available. This research ismotivated in part by the potential application of speaker recog-nition technologies on handheld devices or the Internet. Whilethe technologies promise an additional biometric layer of securityto protect the user, the practical implementation of such systemsfaces many challenges. One of these is environmental noise. Due tothe mobile nature of such systems, the noise sources can be highlytime-varying and potentially unknown. This raises the require-ment for noise robustness in the absence of information about thenoise. This paper describes a method that combines multicondi-tion model training and missing-feature theory to model noisewith unknown temporal-spectral characteristics. Multiconditiontraining is conducted using simulated noisy data with limitednoise variation, providing a “coarse” compensation for the noise,and missing-feature theory is applied to refine the compensationby ignoring noise variation outside the given training conditions,thereby reducing the training and testing mismatch. This paperis focused on several issues relating to the implementation of thenew model for real-world applications. These include the gener-ation of multicondition training data to model noisy speech, thecombination of different training data to optimize the recognitionperformance, and the reduction of the model’s complexity. Thenew algorithm was tested using two databases with simulated andrealistic noisy speech data. The first database is a redevelopmentof the TIMIT database by rerecording the data in the presence ofvarious noise types, used to test the model for speaker identifica-tion with a focus on the varieties of noise. The second database isa handheld-device database collected in realistic noisy conditions,used to further validate the model for real-world speaker verifica-tion. The new model is compared to baseline systems and is foundto achieve lower error rates.

La justice sociale dans la jurisprudence au Royaume-Uni

This chapter surveys the extent to which UK courts have developed the concept of social justice. It focuses on decisions reached in the areas of equality, welfare law, education, and health care, and concludes with a consideration of the extent to which UK judges consider that individuals should take personal responsibility for their own well-being.

Assessment of the effectiveness of the guard ring in obtaining a uni-directional flow in an in situ water permeability test

The non-destructive evaluation of the water permeability of concrete structures is a long standing challenge, principally due to the difficulty of achieving a uni-direction flow for computing the water permeability coefficient. The use of a guard ring (GR) was originally proposed for the in situ sorptivity test, but little information can be found for the water permeability test. In this study, the effect of a GR was carefully examined through the flow simulation, which was verified by carrying out experiments. It was observed that the GR can confine the flow near the surface, but cannot achieve a uni-directional flow across the whole depth of flow. To achieve a better performance, it is essential to consider the effects of the size of the inner seal and the GR and the significant interaction between these two. The analysis of the experimental data has indicated that the GR influences the flow for porous concretes, but there is no significant effect for dense concretes. Further investigation, validated using the flow-net theory, has shown a strong correlation between the water permeability coefficients obtained with the GR (K w-GR) and without it (K w-No GR), suggesting that one dimensional flow is not essential for interpreting data for site tests. Another practical issue was that more than 30 % of the tests with GR failed due to the difficulty of achieving a good seal between the inner and the outer chambers. Based on the work reported in this paper, a new water permeability test is proposed.

Seasonal differences in the effects of oscillatory and uni-directional flow on the growth and nitrate-uptake rates of juvenile Laminaria digitata (Phaeophyceae)

The influence of oscillatory versus unidirectional flow on the growth and nitrate-uptake rates of juvenile kelp, Laminaria digitata, was determined seasonally in experimental treatments that simulated as closely as possible natural environmental conditions. In winter, regardless of flow condition (oscillatory and unidirectional) or water velocity, no influence of water motion was observed on the growth rate of L. digitata. In summer, when ambient nitrate concentrations were low, increased water motion enhanced macroalgal growth, which is assumed to be related to an increase in the rate of supply of nutrients to the blade surface. Nitrate-uptake rates were significantly influenced by water motion and season. Lowest nitrate-uptake rates were observed for velocities <5 cm · s−1 and nitrate-uptake rates increased by 20%–50% under oscillatory motion compared to unidirectional flow at the same average speed. These data further suggested that the diffusion boundary layer played a significant role in influencing nitrate-uptake rates. However, while increased nitrate-uptake in oscillatory flow was clear, this was not reflected in growth rates and further work is required to understand the disconnection of nitrate-uptake and growth by L. digitata in oscillatory flow. The data obtained support those from related field-based studies, which suggest that in summer, when insufficient nitrogen is available in the water to saturate metabolic demand, the growth rate of kelps will be influenced by water motion restricting mass transfer of nitrogen.