How may I help you

http://sisl.disi.unitn.it/wp-content/uploads/2014/11/fragclustering-speechcomm-19981.pdf

Grammar fragment acquisition using syntactic and semantic clustering

The Philips automatic train timetable information system

This article presents a unified approach to Chinese statistical language modeling (SLM). Applying SLM techniques like trigram language models to Chinese is challenging because (1) there is no standard definition of words in Chinese; (2) word boundaries are not marked by spaces; and (3) there is a dearth of training data. Our unified approach automatically and consistently gathers a high-quality training data set from the Web, creates a high-quality lexicon, segments the training data using this lexicon, and compresses the language model, all by using the maximum likelihood principle, which is consistent with trigram model training. We show that each of the methods leads to improvements over standard SLM, and that the combined method yields the best pinyin conversion result reported.

/pdf/toward-a-unified-approach-to-statistical-language-modeling-4c7ugzwa1k.pdf

Toward a unified approach to statistical language modeling for Chinese

In this paper, we describe a set of metrics for the evaluation of different dialogue management strategies in an implemented real-time spoken language system. The set of metrics we propose tries to offer useful insights in evaluating how particular choices in the dialogue management can affect the overall quality of the man-machine dialogue. The evaluation makes use of established metrics: the transaction success, the contextual appropriateness of system answers, the calculation of normal and correction turns in a dialogue. We also define a new metric, the implicit recovery, which allows to measure the ability of a dialogue manager to deal with errors by different levels of analysis. We report evaluation data from several experiments, and we compare two different approaches to dialogue repair strategies using the set of metrics we argue for.

/pdf/metrics-for-evaluating-dialogue-strategies-in-a-spoken-4a6p86peg2.pdf

Metrics for evaluating dialogue strategies in a spoken language system

In some speech recognition tasks, such as man-machine dialogue systems, the spoken sentences include several recurrent phrases. A bigram language model does not adequately represent these phrases because it underestimates their probability. A better approach consists of modeling phrases as if they were individual dictionary elements. They we inserted as additional entries into the word lexicon, on which bigrams are finally computed. This paper discusses two procedures for automatically determining frequent phrases (within the framework of a probabilistic language model) in an unlabeled training set of written sentences. One procedure is optimal since it minimises the set perplexity. The other, based on information theoretic criteria, insures that the resulting model has a high statistical robustness. The two procedures are tested on a 762-word spontaneous speech recognition task. They give similar results and provide a moderate improvement over standard bigrams.

Phrase bigrams for continuous speech recognition

The representation of language models through stochastic finite state networks offers several attractive features for speech recognition and understanding, including the ease of integration with the algorithms used for acoustic-phonetic decoding However, the application to real-world problems is difficult because the network sizes grow very large, and their training requires grammatical inference methods An approach is described that keeps the network size small by avoiding detailed modeling of linguistic constructs not essential for understanding the meaning of sentences An automatic method for training such networks starting from a corpus of semantically annotated sentences is described The training procedure learns both the network structure and the rules for generating the semantic representation of sentences Testing of a 787-word task achieved 92% correct sentence understanding with written input and 72% with continuous speech, speaker-independent, telephone-bandwidth spoken input >

E.P. Giachin

Papers

Phrase bigrams for continuous speech recognition

Automatic training of stochastic finite-state language models for speech understanding