Data Processing and Analysis

Transcribe audio and video

Interviews are usually documented in the form of video or audio. The library offers workshops and guidance in tools to transcribe interviews and to analyse the transcription. 

Transcribe and analyse interviews



ELAN is an open-source transcription tool for audio and video where you can use tiers for different speakers or for speech and gestures. It allows for segmentation and registration of silences and is a good option for e.g. conversation analysis and analysis of sign language.

Automatic transcription

Interviews contain personal data, which limits which software is appropriate to use to transcribe the recording. 

If the interview material does not contain sensitive personal data, a suitable transcription tool is available in Canvas Studio. It is intended for captioning video but can also be used to transcribe audio and video. Keep in mind that the automatic transcription needs to be edited manually as the result of automatic transcription is not flawless. 

There is currently no procured software for transcribing data containing sensitive personal data for Uppsala University. Researchers can apply for projects on UPPMAX and within that project use OpenAI Whisper. As long as Whisper is used at UPPMAX, it can be used for sensitive data. It is also possible for researchers to purchase the Sunet speech-to-text service directly from Amberscript provided that they also sign a personal data processing agreement (DPA/PUBA) with them.