RDM resources and services: KA³ audio services
KA3 Audioservices have been deprecated. For more information, please contact info-dch@uni-keoln.de
Audio and video recordings are made in humanities research as part of various subjects and data collection methods. Among other things, voice recordings are often made of interviews. Evaluating and analyzing such audio recordings, for example through labor-intensive transcripts, is usually time-consuming and data-intensive.
For this reason, most humanities disciplines work with written representations of language. However, the transcription of audio recordings is also time-consuming.
Automatic speech recognition can significantly reduce the time required. The results can vary depending on the language register and recording quality, but even with poorer quality results that require post-processing, a noticeable time saving can usually be achieved.
In the BMBF joint project "KA³ - Cologne Center for Analysis and Archiving of AV Data", the DCH and its project partners Fraunhofer Institute for Intelligent Analysis and Information Systems(IAIS) and the Regional Computing Center of the University of Cologne(RRZK) have set up audio services.
The KA³ audio services offer
- orthographic transcription,
- chapter recognition,
- academic use and
- a simple application.
The web services are offered by the Data Center for the Humanities. The automatic speech recognition was developed by the Fraunhofer IAIS as part of the BMBF joint project KA³ and the services are provided in cooperation with the RRZK.
For optimal and efficient advice and support in the transcription and analysis of audiovisual (AV) data, we are available to you in the current situation by e-mail or video conference. You can reach us easily and conveniently via info-dch(at)uni-koeln.de.
