Speech Recognition with unknown partial feature corruption - a review of the union model

This paper provides a summary of our studies on robust speech recognition based on a new statistical approach – the probabilistic union model. We consider speech recognition given that part of the acoustic features may be corrupted by noise. The union model is a method for basing the recognition on the clean part of the features, thereby reducing the effect of the noise on recognition. To this end, the union model is similar to the missing feature method. However, the two methods achieve this end through different routes. The missing feature method usually requires the identity of the noisy data for noise removal, while the union model combines the local features based on the union of random events, to reduce the dependence of the model on information about the noise. We previously investigated the applications of the union model to speech recognition involving unknown partial corruption in frequency band, in time duration, and in feature streams. Additionally, a combination of the union model with conventional noise-reduction techniques was studied, as a means of dealing with a mixture of known or trainable noise and unknown unexpected noise. In this paper, a unified review, in the context of dealing with unknown partial feature corruption, is provided into each of these applications, giving the appropriate theory and implementation algorithms, along with an experimental evaluation.

Effects of n-3 long chain polyunsaturated fatty acids on depressed mood: Systematic review of published trials.

Background: Greater dietary intakes of n–3 long-chain polyunsaturated fatty acids (n–3 PUFAs) may be beneficial for depressed mood. Objective: This study aimed to systematically review all published randomized controlled trials investigating the effects of n–3 PUFAs on depressed mood. Design: Eight medical and health databases were searched over all years of records until June 2006 for trials that exposed participants to n–3 PUFAs or fish, measured depressed mood, were conducted on human participants, and included a comparison group. Results: Eighteen randomized controlled trials were identified; 12 were included in a meta-analysis. The pooled standardized difference in mean outcome (fixed-effects model) was 0.13 SDs (95% CI: 0.01, 0.25) in those receiving n–3 PUFAs compared with placebo, with strong evidence of heterogeneity (I2 = 79%, P <0.001). The presence of funnel plot asymmetry suggested that publication bias was the likely source of heterogeneity. Sensitivity analyses that excluded one large trial increased the effect size estimates but did not reduce heterogeneity. Metaregression provided some evidence that the effect was stronger in trials involving populations with major depression—the difference in the effect size estimates was 0.73 (95% CI: 0.05, 1.41; P = 0.04), but there was still considerable heterogeneity when trials that involved populations with major depression were pooled separately (I2 = 72%, P <0.001). Conclusions: Trial evidence that examines the effects of n–3 PUFAs on depressed mood is limited and is difficult to summarize and evaluate because of considerable heterogeneity. The evidence available provides little support for the use of n–3 PUFAs to improve depressed mood. Larger trials with adequate power to detect clinically important benefits are required.

Review of Aerospace Engineering Cost Modelling: The Genetic Causal Approach

The primary intention of this paper is to review the current state of the art in engineering cost modelling as applied to aerospace. This is a topic of current interest and in addressing the literature, the presented work also sets out some of the recognised definitions of cost that relate to the engineering domain. The paper does not attempt to address the higher-level financial sector but rather focuses on the costing issues directly relevant to the engineering process, primarily those of design and manufacture. This is of more contemporary interest as there is now a shift towards the analysis of the influence of cost, as defined in more engineering related terms; in an attempt to link into integrated product and process development (IPPD) within a concurrent engineering environment. Consequently, the cost definitions are reviewed in the context of the nature of cost as applicable to the engineering process stages: from bidding through to design, to manufacture, to procurement and ultimately, to operation. The linkage and integration of design and manufacture is addressed in some detail. This leads naturally to the concept of engineers influencing and controlling cost within their own domain rather than trusting this to financers who have little control over the cause of cost. In terms of influence, the engineer creates the potential for cost and in a concurrent environment this requires models that integrate cost into the decision making process.