The transcription, annotation and alignment of audio files is widely used as a research method and is applied at all Unibe language institutes. Until now, data has been transcribed manually in a very time-consuming process. Automated transcription and the subsequent annotation and alignment accelerate and facilitate this process massively, but also harbour risks that students and researchers have to be aware of.
In this Unibe DigiK perspective project, automated transcription will be established at all language institutes and in the digital humanities. Students and researchers will learn and discuss the use of the tools, which will also be trained for non-standard varieties in order to facilitate sociolinguistic research methods.
