We aim to replace the long and complicated, pipeline employed to produce probabilistic phrasal bilexica with a theoretically principled, grammar based, approach. To this end, we introduce a learning regime to learn a phrasal grammar equivalent to linear transduction grammars. The stochastic version of this new grammar type also has the property that the set of biterminals constitute a natural p...

Cepstral Mean Normalization (CMN) is a widely used technique for channel compensation and for noise robustness. CMN compensates for noise by transforming both train and test utterances to zero mean, thus matching first-order moment of train and test conditions. Since all utterances are normalized to zero mean, CMN could lead to loss of discriminative speech information, especially for short utt...

We have developed an approach to natural language processing in which the natural language processor is viewed as a knowledge-based system whose knowledge is about the meanings of the utterances of its language. The approach is orzented around the phrase rather than the word as the basic unit. We believe that this paradi~ for language processing not only extends the capabilities of other natura...

This paper investigates the effects of limited speech data in the context of speaker verification using deep neural network (DNN) approach. Being able to reduce the length of required speech data is important to the development of speaker verification system in real world applications. The experimental studies have found that DNN-senone-based Gaussian probabilistic linear discriminant analysis ...

Computer Assisted Pronunciation Training systems have become popular tools to train on second languages. Many second language learners prefer to train on pronunciation in a stress free environment with no other listeners. There exists no such tool for training on pronunciation of the Norwegian language. Pronunciation exercises in training systems should be directed at important properties in th...

Hiero translation models have two limitations compared to phrase-based models: 1) Limited hypothesis space; 2) No lexicalized reordering model. We propose an extension of Hiero called PhrasalHiero to address Hiero’s second problem. Phrasal-Hiero still has the same hypothesis space as the original Hiero but incorporates a phrase-based distance cost feature and lexicalized reodering features into...

NTCIR-2 evaluation experiments at the Justsystem site are described with a focus on comparative study of CLIR effectiveness with monolingual retrieval effectiveness of the same retrieval engine. Experiments on the effects of phrasal translation, indexing of translated phrasal terms, pre-translation feedback and parallel documents feedback in diverse retrieval settings, are reported. The results...