A new language model is proposed to cope with the demands for recognizing out-of-vocabulary (OOV) words not registered in the lexicon. This language model is a class N-gram incorporating a set of word models that reflect the statistical characteristics of the phonotactics, which depend on the lexical classes. Utilization of class-dependency enhances recognition accuracy and enables identificati...

Decoders that make use of token-passing restrict their search space by various types of token pruning. With use of the Language Model Look-Ahead (LMLA) technique it is possible to increase the number of tokens that can be pruned without loss of decoding precision. Unfortunately, for token passing decoders that use single static pronunciation prefix trees, full n-gram LMLA increases the needed n...

The present article investigates the fusion of different language models to improve translation accuracy. A hybrid MT system, recentlydeveloped in the European Commissionfunded PRESEMT project that combines example-based MT and Statistical MT principles is used as a starting point. In this article, the syntactically-defined phrasal language models (NPs, VPs etc.) used by this MT system are supp...

Automatically generated playlists have become an important medium for accessing and exploring large collections of music. In this paper, we present a probabilistic model for generating coherent playlists by embedding songs and social tags in a unified metric space. We show how the embedding can be learned from example playlists, providing the metric space with a probabilistic meaning for song/s...

Developing better methods for segmenting continuous text into words is important for improving the processing of Asian languages, and may shed light on how humans learn to segment speech. We propose two new Bayesian word segmentation methods that assume unigram and bigram models of word dependencies respectively. The bigram model greatly outperforms the unigram model (and previous probabilistic...

In a task oriented domain, recognizing the intention of a speaker is important so that the conversation can proceed in the correct direction. This is possible only if there is a way of labeling the utterance with its proper intent. One such labeling techniques is Dialog Act (DA) tagging. This work focuses on discussing various n-gram DA tagging techniques. In this paper, a new method is propose...

This work proposes a new research direction to address the lack of structures in traditional n-gram models. It is based on a weakly supervised dependency parser that can model speech syntax without relying on any annotated training corpus. Labeled data is replaced by a few hand-crafted rules that encode basic syntactic knowledge. Bayesian inference then samples the rules, disambiguating and com...

The n-gram model is a stochastic model, which predicts the next word (predicted word) given the previous words (conditional words) in a word sequence. The cluster n-gram model is a variant of the n-gram model in which similar words are classified in the same cluster. It has been demonstrated that using different clusters for predicted and conditional words leads to cluster models that are super...

Adopting concepts from statistical language modeling and rulebased transformations can lead to effective and efficient text-tophone (TTP) functions. We present here the methods and results of one such effort, resulting in a relatively compact and fast set of TTP rules that achieves 94.5% segmental phonemic accuracy.

This paper describes the use of a weighted mixture of classbased n-gram language models to perform topic adaptation. By using a fixed class n-gram history and variable word-given-class probabilities we obtain large improvements in the performance of the class-based language model, giving it similar accuracy to a word n-gram model, and an associated small but statistically significant improvemen...