Hidden Conditional Random Fields for Visual Speech Recognition

In this paper we present the application of Hidden Conditional Random Fields (HCRFs) to modelling speech for visual speech recognition. HCRFs may be easily adapted to model long range dependencies across an observation sequence. As a result visual word recognition performance can be improved as the model is able to take more of a contextual approach to generating state sequences. Results are presented from a speaker-dependent, isolated digit, visual speech recognition task using comparisons with a baseline HMM system. We firstly illustrate that word recognition rates on clean video using HCRFs can be improved by increasing the number of past and future observations being taken into account by each state. Secondly we compare model performances using various levels of video compression on the test set. As far as we are aware this is the first attempted use of HCRFs for visual speech recognition.

Recognition and Redistribution in Theories of Justice Beyond the State

We consider here how cultural and socioeconomic dimensions of justice beyond the state are related. First we examine cosmopolitan theories that have drawn on John Rawls's egalitarian liberal framework to argue that a just global order requires substantive, transnational redistribution of material resources. We then assess the view, ironically put forward by Rawls himself, that this perspective is ethnocentric and insufficiently tolerant of non-liberal cultures. We argue that Rawls is right to be concerned about the danger of ethnocentrism, but wrong to assume that this requires us to reject the case for substantive redistribution across state boundaries. A more compelling account of justice beyond the state will integrate effectively socioeconomic and cultural aspects of justice. We suggest that this approach is best grounded in a critical theory of recognition that responds to the damage caused to human relations by legacies of historical injustice.