We describe how language models, combined with models of pen placement, can be used to significantly reduce the error rate of soft keyboard usage, by allowing for cases in which a key press is outside of a key boundary. Language models predict the probabilities of words or letters. Soft keyboards are images of keyboards on a touch screen used for input on Personal Digital Assistants. When a soft keyboard user hits a key near the boundary of a key position, we can use the language model and key press model to select the most probable key sequence, rather than the sequence dictated by strict key boundaries. This leads to an overall error rate reduction by a factor of 1.67 to 1.87.

This paper investigates the perplexity and word error rate performance of two different forms of class model and the respective data-driven algorithms for obtaining automatic word classifications. The computational complexity of the algorithm for the `conventional' two-sided class model is found to be unsuitable for very large vocabularies ( 100k) or large numbers of classes ( 2000). A one-sided class model is therefore investigated and the complexity of its algorithm is found to be substantially less in such situations. Perplexity results are reported on both English and Russian data. For the latter both 65k and 430k vocabularies are used. Lattice rescoring experiments are also performed on an English language broadcast news task. These experimental results show that both models, when interpolated with a word model, perform similarly well. Moreover, classifications are obtained for the one-sided model in a fraction of the time required by the two-sided model, especially for very large vocabularies.

Building probabilistic models of language is a central task in natural language and speech processing allowing to integrate the syntactic and/or semantic (and recently pragmatic) constraints of the language into the systems. Probabilistic language models are an attractive alternative to the more traditional rule-based systems, such as context free grammars, because of the recent availability of massive amount of text corpora which can be used to e#ciently train the models and because instead of binary grammaticality judgement o#ered by the rule-based systems, likelihood of any sequence of lexical units can be obtained, which is a crucial factor in such tasks as speech recognition. Probabilistic language models also find their application in part-of-speech tagging, machine translation, semantic disambiguation and numerous other fields.

This thesis addresses the application of automatic speech recognition to the task of offline closed-captioning of television programs, and describes the collection of corpora to support such research and an exploration of issues to be addressed. The use of automatic speech recognition (ASR) for transcription of broadcast speech and as an aid to captioning is reviewed. As background to the task, the methodology for large vocabulary continuous speech recognition (LVCSR) is presented, with particular attention given to the issues of large vocabulary language modelling and consideration of the acoustic complexity arising in broadcast material. A speech corpus of segmented and transcribed speech utterances for ten program episodes was developed for a typical genre of television programming (travelogues) for which offline closed-captions are applied. The development of this corpus demonstrates the feasibility of using existing closed-caption sources for generating labelled acoustic data suitable for speech recognition research. The speech corpus exhibits far greater acoustic complexity and much lower signal to noise ratios than occurs in broadcast news data (which has been systematically evaluated in ASR research). Noise-tolerant speech recognisers were developed and effectively

In this paper we present a method for document organization and retrieval based on statistical language modeling. The proposed method, which is based on the vector model, uses nonlinear interpolation to provide more accurate statistical estimators of the conditional probabilities employed for encoding the context of each word. An information retrieval system is built using the self-organizing map algorithm. In the first step, the self-organizing architecture is used to cluster the feature vectors and to build clusters of semantically related words. Subsequently, the collection of documents is encoded into vectors and the same algorithm is used to cluster the documents in contextually related classes

We present two language models based upon an "immediate-head" parser --- our name for a parser that conditions all events below a constituent c upon the head of c. While all of the most accurate statistical parsers are of the immediate-head variety, no previous grammatical language model uses this technology. The perplexity for both of these models significantly improve upon the trigram model baseline as well as the best previous grammar-based language model. For the better of our two models these improvements are 24% and 14% respectively. We also suggest that improvement of the underlying parser should significantly improve the model's perplexity and that even in the near term there is a lot of potential for improvement in immediate-head language models.

take two forms. Human input can be constrained through a directed dialog, allowing the decoder to use a state-specific language model to improve recognition accuracy. Mixedinitiative systems allow for human input that while domainspecific might not be state-specific. Nevertheless, for the most part human input to a mixed-initiative system is predictable, particularly when given information about the immediately preceding system prompt. The work reported in this paper addresses the problem of balancing state-specific and general language modeling in a mixed-initiative dialog system. By incorporating dialog state adaptation of the language model, we have reduced the recognition error rate by 11.5%.