A fast and portable realizer for text generation systems

Lavoie, Benoit; Rainbow, Owen

doi:10.3115/974557.974596

Cited by 127 publications

(92 citation statements)

References 6 publications

(4 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The top ranked sentence plan output by the SPR is input to the RealPro surface realizer which produces a surface linguistic utterance (Lavoie and Rambow, 1997). A prosody assignment module uses the prior levels of linguistic representation to determine the appropriate prosody for the utterance, and passes a markedup string to the text-to-speech module.…”

Section: 70mentioning

confidence: 99%

Trainable sentence planning for complex information presentation in spoken dialog systems

Stent

Prassad

Walker

2004

Proceedings of the 42nd Annual Meeting on Association for Computational Linguistics - ACL '04

106

103

View full text Add to dashboard Cite

A challenging problem for spoken dialog systems is the design of utterance generation modules that are fast, flexible and general, yet produce high quality output in particular domains. A promising approach is trainable generation, which uses general-purpose linguistic knowledge automatically adapted to the application domain. This paper presents a trainable sentence planner for the MATCH dialog system. We show that trainable sentence planning can produce output comparable to that of MATCH's template-based generator even for quite complex information presentations.

show abstract

Section: 70mentioning

confidence: 99%

Trainable sentence planning for complex information presentation in spoken dialog systems

Stent

Prassad

Walker

2004

Proceedings of the 42nd Annual Meeting on Association for Computational Linguistics - ACL '04

106

103

View full text Add to dashboard Cite

show abstract

“…Such a deep-syntactic tree facilitates the manageable yet comprehensive storage of the constituents of a sentence. In addition, it can be automatically mapped to a syntactically correct sentence with existing tools [24]. Taking the example of the activity Choose Contact Type, the corresponding DSynT consists of a root node pointing to the verb choose and two subordinate nodes.…”

Section: Dsynt-message Generationmentioning

confidence: 99%

“…Existing tools significantly vary in aspects such as license costs, generation speed, and Java compatibility. Taking these aspects into account, we decided to use the DSynT-based realizer RealPro from CoGenTex [24]. RealPro requires an XML-based DSynT message as input and transforms it to a grammatically correct sentence.…”

Section: Dsynt-message Generationmentioning

confidence: 99%

Creating and Updating Personalized and Verbalized Business Process Descriptions

Kolb¹,

Leopold

Mendling

et al. 2013

Lecture Notes in Business Information Processing

View full text Add to dashboard Cite

Abstract. The increasing adoption of process-aware information systems (PAISs) has resulted in large process model collections. To support users having different perspectives on complex processes and related data, a PAIS should enable personalized process views, i.e., user-specific abstractions of process models. Despite the abstraction achieved through views of the graphical process models, many end users still struggle with understanding these graphical models and their details. For selected user groups, therefore, a PAIS should provide verbalized process descriptions describing their role in the process. Existing PAISs neither provide mechanisms for managing process views nor verbalized process descriptions. While process views have been used as visual abstractions for large process models, so far no work exists on how to provide both personalized and verbalized process descriptions based on respective views. This paper presents an approach for creating such personalized and verbalized process descriptions based on process views. Furthermore, textual changes of a personalized and verbalized process description are correctly mapped to corresponding updates of the underlying process model. In this context, all other views and process descriptions related to this process model are migrated to the new version of the process model as well. Overall, our approach enables end users to understand and evolve large process models based on personalized and verbalized process descriptions.

show abstract

“…The main challenge in using NLG for knowledge publishing, therefore, is to develop tools and techniques that will enable knowledge engineers, instead of linguists, to create and customise these resources. The system development costs will also be reduced by re-using efficient NLG components such as RealPro (Lavoie 1997) and frameworks such as Exemplars (White and Caldwell 1998).…”

Section: Knowledge Publishingmentioning

confidence: 99%

Using HLT for acquiring, retrieving and publishing knowledge in AKT

Bontcheva

Brewster

Ciravegna

et al. 2001

Proceedings of the Workshop on Human Language Technology and Knowledge Management -

View full text Add to dashboard Cite

AKT is a major research project applying a variety of technologies to knowledge management. Knowledge is a dynamic, ubiquitous resource, which is to be found equally in an expert's head, under terabytes of data, or explicitly stated in manuals. AKT will extend knowledge management technologies to exploit the potential of the semantic web, covering the use of knowledge over its entire lifecycle, from acquisition to maintenance and deletion. In this paper we discuss how HLT will be used in AKT and how the use of HLT will affect different areas of KM, such as knowledge acquisition, retrieval and publishing.Research Memoranda CS-01-14 AbstractAKT is a major research project applying a variety of technologies to knowledge management. Knowledge is a dynamic, ubiquitous resource, which is to be found equally in an expert's head, under terabytes of data, or explicitly stated in manuals. AKT will extend knowledge management technologies to exploit the potential of the semantic web, covering the use of knowledge over i ts entire lifecycle, from acquisition to maintenance and deletion. In this paper we discuss how HLT will be used in AKT and how the use of HLT will affect different areas of KM, such as knowledge acquisition, retrieval and publishing.

show abstract

A fast and portable realizer for text generation systems

Cited by 127 publications

References 6 publications

Trainable sentence planning for complex information presentation in spoken dialog systems

Trainable sentence planning for complex information presentation in spoken dialog systems

Creating and Updating Personalized and Verbalized Business Process Descriptions

Using HLT for acquiring, retrieving and publishing knowledge in AKT

Contact Info

Product

Resources

About