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.

An Investigation into the Link between UK Credit Union Characterisitics, Location and their Success

The unique characteristics of credit unions reduces the information asymmetry that is prevalent in credit making decisions, enabling them to provide loans where other financial institutions cannot. This makes them a potential tool in the fight against financial exclusion. Yet, the UK credit union movement is not regarded as being successful, even though there is evidence of much financial exclusion. This study is cross sectional in form, and evaluates characteristics that may contribute to the success of the UK credit union movement at national and regional level, in 2000. The findings are used to consider the impact of recent regulatory changes on the movement. The key findings are that there is a significant relationship between the success of a credit union, its size and the deprivation of the ward from which it sources its members. More specifically, larger credit unions and those located in more affluent wards, are more successful. Affiliation to the Irish League of Credit Unions and having a common bond of occupation, are also found to be contributing factors to credit union success. These results are taken as providing support for the recent changes implemented by the Financial Services Authority (FSA), which is likely to result in the emergence of larger credit unions (through mergers), run by appropriately qualified persons, serving a more mixed-income membership base. It is, however, noted that the history of the UK movement is one of missed opportunities and only time will tell whether credit unions have the wherewithal to accept current opportunities.

The trade unions membership dynamics and union members’ trajectories

This text presents an analysis of aggregated membership’s dynamics for Spanish trade unions, using ECVT data, as well as union memberships’ trajectories, or members’ decisions about joining the organization, permanency and responsibilities, and subsequent attrition. For the analysis of trajectories we make use of information of the records of actual memberships and the record of quitting of CCOO, and of a survey-questionnaire to a sample of leavers of the same union. This study allows us to confirm a linkage between the decision and motivations to become union member, to participate in union activities, the time of permanency, and the motives to quit the organization. We also identify five types of union members’ trajectories, indicating that, far from views that assert a monolithic structure, unions are complex organizations.