Processing speech acts: spoken communication for aircraft maintenance
DOI:
https://doi.org/10.36505/ExLing-2023/14/0007/000601Keywords:
Spoken technical texts, Speech acts, Template-based slot-filling framework, Commands, Aircraft maintenance, NLUAbstract
The identification of speech act types and speech act content constitutes the basis for the construction and implementation of a flexible template-based, slot-filling framework for spoken technical texts in the domain of aircraft maintenance for its direct deployment or its subsequent integration in neural network approaches and/or as training data. Keywords extracted from the content of each spoken utterance (620 expressions, approximately 2800 spoken utterances) are integrated into the “generic-intent” slot of the template-based, slot-filling framework, which is, in turn, linked to the respective speech act. The mandatory recognition of two keywords in complementary relation (x,y) within the framework is observed to produce correct results for Greek spoken technical texts.
References
Jurafsky, D., Martin, J. H. 2022 (3rd ed.). Speech and Language Processing: An Introduction to Natural Language Processing, Computational Linguistics and Speech Recognition.
Lehrndorfer, A. 1996. Kontrolliertes Deutsch: Linguistische und Sprachpsychologische Leitlinien für eine (maschinell) kontrollierte Sprache in der technischen Dokumentation. Tübingen, Germany: Gunter Narr Verlag.
Okur, E., Kumar, S. H., Sahay, S., et al. 2023. Natural Language Interactions in Autonomous Vehicles: Intent Detection and Slot Filling from Passenger Utterances.
Wojcik, R. H., Holmback, H. 1996. Getting a controlled language off the ground at Boeing. In Proceedings of CLAW–1996, 22-31. Leuven, Belgium.
Downloads
Published
Issue
Section
License
Articles are published under the Creative Commons Attribution 4.0 International License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.
