NFDI4Memory is looking for pilot users for automated transcription service

The project ASR4Memory project is looking for pilot users for the automated transcription of audiovisual research data in historical contexts.

The NFDI4Memory-funded project is developing a prototype service that enables the automated transcription of historical audiovisual resources from heterogeneous sources in different languages for various research, re-use and archiving scenarios. For this, the data is prepared qualitatively, processed by open source-based speech recognizers and finally converted into time-coded transcript and metadata formats according to scientific standards. In addition, the speech recognition models are further trained and fine-tuned.

Pilot users provide their audiovisual data temporarily. These are processed locally and securely and deleted at the end of the project. The users receive the generated transcriptions in various export formats for their further use.

Further information and participationExternal link

NFDI4MemoryExternal link is one of the 27 consortia of the National Research Data Infrastructure (NFDI), which jointly develop sustainable solutions and services for the management of research data. NFDI4Memory brings together all stakeholders who use historical research methods or work with data in a historical context in order to jointly address their specific challenges.