| Levin, L.; Thyme-Gobbel, A,; Lavie, A.; Ries, K.; Zechner, K., 1998. A Discourse Coding Scheme for Conversational Spanish. In International Conference on Spoken Language Processing (ICSLP '98). |
....on three discourse levels: speech acts, dialogue games, and functional activities. Tagging, training, and testing was carried out. on the Call Home Spanish (GHS) corpus of casual, non task oriented conversation between family members. The coding scheme for speech acts and games is described iu [6, 9]. The automatic classification of speech acts and games is carried out in a multi level architecture that integrates classification at multiple discourse levels instead of performing them sequentially. The multi level classifier is described in [7] The focus of this paper is on the cross level ....
....act coding is based on the SWBDDAMSL tag set from the Johns Hopkins LVCSR Summer Workshop 1997 [5] Modifications were made in the tag set for four reasons. First, we made some changes to improve inter coder agreement. After two taggers tagged CHS, we calculated intercoder agreement (79 ) See [6] for a description of how this was computed) and examined confusion matrices showing the tags that they disagreed on most. We eliminated the tag sv for statement of opinion because taggers did not distinguish reliably between sv and s, plain statement. To account for some aspects of opinions that ....
[Article contains additional citation context not shown here]
Lori Levin, Ann Thym-Gobbel, Klaus Ries, Alon Lavie, and Monika Woszczyna. A discourse coding scheme for conversational span- ish. In ICSLP, 1998.
....on the CallHome Spanish database. CallHome Spanish consists of 120 telephone conversations of Spanish speakers calling from the US back to their home country. Calls are usually 30 minutes long. We have developed annotation schemes for three levels of discourse structure. In previous publications (Levin et al. 1998; Ries, 1999a) we reported on our annotation of speech acts and dialogue games and the technology used to automatically identify and classify these levels of discourse structure. Work by collaborators in CLARITY included emotion detection and summarization. The focus in this paper is our recent ....
....Using vanilla speech acts (traditional DAMSLspeech acts (Core and Allen, 1997; Jurafsky et al. 1997; Stolcke et al. 1998) we did not see any improvements. However, when we used the enhancements in our tagging scheme such as as future statements, value judgments, certainty and hypotheticals (Levin et al. 1998; Thyme Gobbel and Levin, 1998) we saw improvements from speech acts. Currently we are also preparing the use of prosodic features such as pitch and power contours into the feature set. The annotation results reported below have been obtained on just 520 activity segments corresponding to 40 ....
[Article contains additional citation context not shown here]
Levin, Lori, Ann Thyme-Gobbel, Klaus Ries, Alon Lavie, and Monika Woszczyna, 1998. A discourse coding scheme for conversational spanish. In ICSLP.
....on three discourse levels: speech acts, dialogue games, and functional activities. Tagging, training, and testing was carried out on the Call Home Spanish (CHS) corpus of casual, non task oriented conversation between family members. The coding scheme for speech acts and games is described in [6, 9]. The automatic classification of speech acts and games is carried out in a multi level architecture that integrates classification at multiple discourse levels instead of performing them sequentially. The multi level classifier is described in [7] The focus of this paper is on the cross level ....
....act coding is based on the SWBDDAMSL tag set from the Johns Hopkins LVCSR Summer Workshop 1997 [5] Modifications were made in the tag set for four reasons. First, we made some changes to improve inter coder agreement. After two taggers tagged CHS, we calculated intercoder agreement (79 . See [6] for a description of how this was computed) and examined confusion matrices showing the tags that they disagreed on most. We eliminated the tag sv for statement of opinion because taggers did not distinguish reliably between sv and s, plain statement. To account for some aspects of opinions that ....
[Article contains additional citation context not shown here]
Lori Levin, Ann Thym'e-Gobbel, Klaus Ries, Alon Lavie, and Monika Woszczyna. A discourse coding scheme for conversational spanish. In ICSLP, 1998.
No context found.
LTGR + 98. Lori Levin, Ann Thyme-Gobbel, Klaus Ries, Alon Lavie, and Monika Woszczyna. A discourse coding scheme for conversational spanish. In ICSLP, 1998.
No context found.
Lori Levin, Ann Thyme-Gobbel, Klaus Ries, Alon Lavie, and Monika Woszczyna. A discourse coding scheme for conversational spanish. In ICSLP, 1998.
No context found.
Levin, L.; Thyme-Gobbel, A,; Lavie, A.; Ries, K.; Zechner, K., 1998. A Discourse Coding Scheme for Conversational Spanish. In International Conference on Spoken Language Processing (ICSLP '98).
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC