Utilisation of Audio Mining Technologies for Researching Public Communication on Multimedia Platforms
Michael Eble & Daniel Stein
DOI 10.17174/dcr.v2.14 (SSOAR)
Abstract: The number and volume of spoken language corpora which are generally available for research purposes increase significantly. That is due to the wide adoption of audio-visual communication on news websites and social web platforms. The respective messages that are published by professional and individual communicators are subject to online content analysis. To date, such analyses strongly rely on manually operated processes which come along with a huge effort for transcribing spoken language corpora into textual content. Hence, challenges like the ever increasing volume, velocity and variety of multimedia content need to be faced. Audio Mining technologies are capable of reducing the effort for turning speech into text significantly. Using these technologies via application programming interfaces (APIs), it is demonstrated how a hybrid approach enables researchers to reduce the time that is needed for analysing news content by an order of magnitude.
Dr. Michael Eble is Consultant at mm1 Consulting & Management PartG in Stuttgart
Dr. Daniel Stein is Senior Research Scientist at the Fraunhofer Institute for Intelligent Analysis and Information Systems IAIS in Sankt Augustin
Eble, M., & Stein, D. (2015). Utilisation of Audio mining Technologies for Researching Public Communication on Multimedia Platforms. In A. Maireder, J. Ausserhofer, C. Schumann, & M. Taddicken (Eds.), Digitale Methoden in der Kommunikationswissenschaft (pp. 329-345). doi: 10.17174/dcr.v2.14
This document is published Open Access and licensed under Creative Commons Attribution 4.0 (CC-BY 4.0).
For persistent archiving we cooperate with the Social Science Open Access Repository and the Registration agency for social and economic data da|ra.