disagreement in conversational speech

... several empirical analyses in or- der to determine to what extent contextual informa- tion helps in discriminating between agreement and disagreement. By integrating the interpretation of the pragmatic ... monitor- ing levels of agreement during a conversation. In Proc. of the 26th Penn Linguistics Colloquium. M. Collins. 2000. Discriminative reranking for nat- ural language parsing. In Proc. 17th Interna- tional ... Entropy Ranking We view the problem as an instance of statisti- cal ranking, a general machine learning paradigm used for example in statistical parsing (Collins, 2000) and question answering (Ravichandran...

Ngày tải lên: 20/02/2014, 16:20

... CRF:All in Table 5). We get about 0.5% increase in accuracy, 76.1% with a window of size w = 1. Using larger windows resulted in minor increases in the performance of the model, as summarized in Table ... 17th International Conference on Machine Learning. A. McCallum. 2003. Efficiently inducing features of Conditional Random Fields. In Proc. of Un- certainty in Articifical Intelligence. T. Minka. ... understanding. In this paper we investigate probabilistic, contex- tual, and phonological factors that in uence pitch accent placement in natural, conversational speech in a sequence labeling setting....

Ngày tải lên: 08/03/2014, 04:22

... SPEECH Imperatives Direct speech Indirect speech Affirmative Imperative V 1 + …… Mike said to Henry, “Come in, please.” S+ told / asked B + to + V 1 + …… Mike told Mary to come in. Negative Imperative Don’t ... Last week / year → The next/ following day → The day before → The following week/ year → The previous week / year B. Practice: I. Change into reported speech: (Imperatives: Câu mệnh lệnh) 1. ... She said to them, “Can you go fishing with me?” III. Change into reported speech: (Advice: Câu khuyên nhủ) 11.The doctor said to the patient, “You should stay in bed today.” 12.The teacher...

Ngày tải lên: 08/07/2013, 01:27

Ngày tải lên: 18/12/2013, 10:08

Ngày tải lên: 18/12/2013, 18:02

... training data and the ngram-based model was retrained on the remaining subset. Figure 2: Empirical differences in sociolinguistic features for Gender on the Switchboard corpus 6 Incorporating ... speakers used for training and 100 speakers used for testing, re- sulting in a total of 4062 conversation sides for training and 808 conversation sides for testing. 4 Modeling Gender via Ngram ... conversation, resulting in the same number of training/test speakers as conver- sations, and thus there was no overlap in speak- ers/partners between training and test sets. Only non-lexical sociolinguistic...

Ngày tải lên: 20/02/2014, 07:20

... train- ing set by either oversampling the minority class or downsampling the majority class. In this prelimi- nary study of sampling approaches for handling im- balanced data for CRF training, ... using constraint classifiers. In Proceedings of HLT/NAACL. D. Hillard, M. Ostendorf, and E. Shriberg. 2003. De- tection of agreement vs. disagreement in meetings: Training with unlabeled data. In ... features using a linear-chain conditional ran- dom fields model and conducted supervised train- ing. We observed significant improvement from adding prosodic features and employing two sam- pling approaches,...

Ngày tải lên: 07/03/2014, 22:20

... The decoding pro- cess proceeds as follows: 3.1 Training and decoding Training data for ALTERF is supplied in the form of a text file containing one example per line, in the format 1. Initialise ... features. • Quantity of training data. • Modality (text or speech) of training data. We randomly divided the training corpus into ten equal pieces, and trained on subsets ranging from 10% of the corpus ... Percentage semantic interpretation errors on in- domain test data for different amounts of training data and different versions of the system. Training and test data both in speech form. "Data" =...

Ngày tải lên: 08/03/2014, 21:20

... utterances used in training. In some experiments (#1 and 2) we trained with the entire training set 3 , including sen- tences without speaker errors, and in others (#3-6) we trained only on those ... Jelinek. 2008. Linguis- tic resources for reconstructing spontaneous speech text. In Proceedings of the Language Resources and Evaluation Conference, May. Erin Fitzgerald. 2009. Reconstructing ... ig- nored for these error tagging experiments. We approach our training of CRFs in several ways, detailed in Table 3. In half of our exper- iments (#1, 3, and 4), we trained a single model to predict...

Ngày tải lên: 08/03/2014, 21:20

... examined in principle, but without going too deeoly into the code. 3.1 Sentence Processing Firstly, the user types in a sentence in normal English text, with word boundaries ind/cated in ... that an increase in intensity with no corresponding pitch increase was never- theless heard as a pitch raise. Interestingly enough, a drop in intensity was not heard as a drop in pitch, merely ... This is indeed true in a great number of cases, but by no means all, as pointed out by Bolinger (Bolinger 1958). For instance, consider the following phrase (taken to mean "do continue"):...

Ngày tải lên: 09/03/2014, 01:20

... prediction in conversational speech. In IWCS 6, Sixth International Workshop on Computational Se- mantics, Tilburg, Netherlands. H Schmid. 1994. Probabilistic part-of -speech tagging using decision ... information. In this paper the best performing measures from (Pucher, 2005), which outperform baseline models on word prediction for conversational tele- phone speech are used for Automatic Speech ... taxonomy. In Proceedings of the International Conference on Re- search in Computational Linguistics, Taiwan. Ted Pedersen, S. Patwardhan, and J. Michelizzi. 2004. WordNet::Similarity - Measuring the...

Ngày tải lên: 17/03/2014, 04:20

... Modelling and Adaptivity in a Speech- based E-mail System Kristiina JOKINEN University of Helsinki and University of Art and Design Helsinki Hämeentie 135C 00560 Helsinki kjokinen@uiah.fi ... speech- based dialogue systems, and to familiarize themselves to synthesized speech and speech recognizers, they had a short training session with another speech application in the beginning ... user expertise. While not entering into discussions about the limits of rule-based thinking (e.g. in order to model intuitive decision making of the experts according to the Dreyfus model), we...

Ngày tải lên: 23/03/2014, 19:20

Ngày tải lên: 24/03/2014, 02:20

Ngày tải lên: 17/04/2014, 14:50

Ngày tải lên: 03/05/2014, 20:50

Ngày tải lên: 20/06/2014, 16:20

Ngày tải lên: 21/06/2014, 16:20

Ngày tải lên: 21/06/2014, 22:20

Ngày tải lên: 22/06/2014, 23:20

Ngày tải lên: 22/06/2014, 23:20