Báo cáo khoa học: "Intonational Boundaries, Speech Repairs and Discourse Markers: Modeling Spoken Dialog" pdf

... Intonational Boundaries, Speech Repairs and Discourse Markers: Modeling Spoken Dialog Peter A. Heeman and James F. Allen Department of Computer Science ... and speech re- pairs. The speech repairs in the Trains corpus have been hand-annotated. We have divided the repairs into three types: fresh starts, modification repairs, and abridged repairs. ... In...

Ngày tải lên: 17/03/2014, 23:20

... producing a positive examples for the part of speech the word is labeled with, and the nega- tive examples for the other parts of speech. The positive and negative examples are presented to the ... dimensionality is very high and that both the learned concepts and the instances reside very sparsely in the feature space. In this paper we present a learning algo- rithm and...

Ngày tải lên: 23/03/2014, 19:20

... word/one preceding word and two following words), and the first and last letter of the word immediately following. (In Fig. 2, “t” and “e” in “telephone”.) The word classes and root forms were ... one preceding/ the targeted word and the one following/ the targeted word and the two preceding/ the targeted word and the two fol- lowing), and the first and last letters...

Ngày tải lên: 20/02/2014, 16:20

... parsing. 1. INTRODUCTION In attempting to merge speech recognition and natural language understanding to produce a system capable of understanding spoken dia- logues, we are confronted with a ... ical and structural ambiguities in spoken English. In this paper we present some results of employ- ing two types of prosodic information, namely pitch and pause, to assist syntac...

Ngày tải lên: 20/02/2014, 21:20

... sentence. (Cohn and La- pata, 2008) expands the operation set by including insertion, substitution and reordering, and incorpo- rates grammar rules. In speech domain, (Clarke and Lapata, 2008) ... trigrams of words and POS tags in the candidate sentence. • Bigrams and trigrams of words and POS tags in the original sentence in combination with their binary labels in the candida...

Ngày tải lên: 07/03/2014, 18:20

... understand and generate natural language, needs knowledge about the real world. As the manual modeling and main- tenance of those knowledge structures, i.e. ontolo- gies, are both time and cost ... the speech recognizer. 4.1 Speech Recognition The speech recognizer classifies all words of the user’s utterance not found in the lexicon as out- 3 In case the speech recognizer of t...

Ngày tải lên: 23/03/2014, 18:20

... also applies only for the relations infer and justify. cue-word inserts a discourse connective (one of since, however, while, and, but, and on the other hand), between the two clauses to be combined. ... decent decor and good service. 3 .67 20 Above and Carmine’s offer exceptional value among the selected restaurants. Carmine’s has decent decor but Above has good decor, and Carmin...

Ngày tải lên: 23/03/2014, 19:20

... user-created content, and a flurry of recent research has aimed to under- stand and exploit these data (Ritter et al., 2010; Shar- ifi et al., 2010; Barbosa and Feng, 2010; Asur and Huberman, 2010; ... our data and provides an appropriate level of linguistic detail, and (2) a feature set that captures Twitter-specific properties and exploits ex- isting resources such as tag diction...

Ngày tải lên: 20/02/2014, 04:20

... (denoted + in the example) between the abandoned thought and its replacement, 2. Determination that the abandoned portion contains unique and preservable content and should be made a new sentence rather ... errors: a leading coordinator (#1 and 2) and rules proceeded by unnecessary filler words (#3 and 5). Complementary rules #7 and 9 (e.g. VP → [rule]/[rule SBAR] and VP → [rul...

Ngày tải lên: 20/02/2014, 07:20

... Workshop on Automatic Speech Recognition and Understand- ing (ASRU), pages 347–354. C. F¨ugen, M. Kolss, D. Bernreuther, M. Paulik, S. St¨uker, S. Vogel, and A. Waibel. 2006. Open domain speech recognition ... translation: Lectures and speeches. In Proc. IEEE Conf. on Acoustics, Speech, and Signal Processing (ICASSP), volume 1, pages 569–572. J. Glass, T.J. Hazen, S. Cyphers, I....

Ngày tải lên: 20/02/2014, 07:20