Automatic detection of discourse structure for speech recognition and understanding

Jurafsky, Daniel; Bates, Rebecca; Coccaro, Noah; Martin, Rachel W.; Meteer, Marie; Ries, Klaus; Shriberg, Elizabeth; Stolcke, Andreas; Taylor, Paul; Ess-Dykema, Carol Van

doi:10.1109/asru.1997.658992

Cited by 85 publications

(82 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Words Inflected Form The word form is used as a baseline lexical feature in most modern lexicalized natural language processing approaches [11,44,32,33]. In our case, sentence segmentation is known but capitalization of the first word of the sentence is removed, which decreases the total number of features in our model without impacting accuracy, thanks to the insertion of a special "start-of-utterance" word.…”

Section: Baseline Featuresmentioning

confidence: 99%

See 1 more Smart Citation

Automatic dialogue act recognition with syntactic features

Král

Cerisara

2014

Lang Resources & Evaluation

View full text Add to dashboard Cite

This work studies the usefulness of syntactic information in the context of automatic dialogue act recognition in Czech. Several pieces of evidence are presented in this work that support our claim that syntax might bring valuable information for dialogue act recognition. In particular, a parallel is drawn with the related domain of automatic punctuation generation and a set of syntactic features derived from a deep parse tree is further proposed and successfully used in a Czech dialogue act recognition system based on Conditional Random Fields. We finally discuss the possible reasons why so few works have exploited this type of information before and propose future research directions to further progress in this area.

show abstract

Section: Baseline Featuresmentioning

confidence: 99%

“…Some cue words and phrases can also serve as explicit indicators of dialogue structure [43]. For example, 88.4% of the trigrams "<start> do you" occur in English in yes/no questions [44].…”

Section: Related Workmentioning

confidence: 99%

Automatic dialogue act recognition with syntactic features

Král

Cerisara

2014

Lang Resources & Evaluation

View full text Add to dashboard Cite

show abstract

“…2 Related work Jurafsky et al (1997a) and Reithinger and Klesen (1997) used n-gram language modeling on the Switchboard and Verbmobil corpora respectively to classify dialog acts. Grau et al (2004) uses a Bayesian approach with n-grams to categorize dialog acts.…”

Section: Introductionmentioning

confidence: 99%

Automatic Identification of Rhetorical Questions

Bhattasali¹,

Cytryn²,

Feldman³

et al. 2015

Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Confere

View full text Add to dashboard Cite

A question may be asked not only to elicit information, but also to make a statement. Questions serving the latter purpose, called rhetorical questions, are often lexically and syntactically indistinguishable from other types of questions. Still, it is desirable to be able to identify rhetorical questions, as it is relevant for many NLP tasks, including information extraction and text summarization. In this paper, we explore the largely understudied problem of rhetorical question identification. Specifically, we present a simple n-gram based language model to classify rhetorical questions in the Switchboard Dialogue Act Corpus. We find that a special treatment of rhetorical questions which incorporates contextual information achieves the highest performance.

show abstract

“…Early approaches start with using the language models [15,16], and also include the use of generative models such as the source-channel model [17], hidden Markov models (HMM) [18,19,20], and the hidden vector state model [21]. Even though discriminative models do not model the joint distribution of features and labels, it is known that they often outperform generative models in classification tasks, since they relax the independence assumption, and enable arbitrary features to be included in the model.…”

Section: Introductionmentioning

confidence: 99%

An empirical investigation of sparse log-linear models for improved dialogue act classification

Chen

Wang

Rudnicky

2013

2013 IEEE International Conference on Acoustics, Speech and Signal Processing

View full text Add to dashboard Cite

Previous work on dialogue act classification have primarily focused on dense generative and discriminative models. However, since the automatic speech recognition (ASR) outputs are often noisy, dense models might generate biased estimates and overfit to the training data. In this paper, we study sparse modeling approaches to improve dialogue act classification, since the sparse models maintain a compact feature space, which is robust to noise. To test this, we investigate various element-wise frequentist shrinkage models such as lasso, ridge, and elastic net, as well as structured sparsity models and a hierarchical sparsity model that embed the dependency structure and interaction among local features. In our experiments on a real-world dataset, when augmenting N -best word and phone level ASR hypotheses with confusion network features, our best sparse log-linear model obtains a relative improvement of 19.7% over a rule-based baseline, a 3.7% significant improvement over a traditional non-sparse log-linear model, and outperforms a state-of-theart SVM model by 2.2%.

show abstract

Automatic detection of discourse structure for speech recognition and understanding

Cited by 85 publications

References 15 publications

Automatic dialogue act recognition with syntactic features

Automatic dialogue act recognition with syntactic features

Automatic Identification of Rhetorical Questions

An empirical investigation of sparse log-linear models for improved dialogue act classification

Contact Info

Product

Resources

About