Getting started with Autotekst

1. Log in to https://autotekst.uio.no.
2. Drag and drop the files you want to transcribe into the upload window, or click to select the file from your device. The following formats are supported: MP3, FLAC, WAV, M4A, MP4, MOV, WEBM. The maximum file size is 4 gigabytes.
3. Choose the language spoken in the recording from the Language menu.
4. Select which language model you want to use for transcription. You can choose between NB Whisper, NB Whisper verbatim, OpenAI Whisper V3, or OpenAI Whisper V2.
5. Additional options available in Autotekst:
   • If the recording has multiple speakers, you can choose to separate them in the transcription by turning on Diarization.
   • If you want to translate the transcription into English, turn on Translate transcription to English. You only receive the translated transcription. If you want the original as well, it must be transcribed separately.
6. Click Send to texting. The file is now placed in the queue. The queue time depends on the size of the file and the demand in Autotekst. Transcription of your queued files will continue even if you leave the page. There are no restrictions on the number of files in the queue.
7. Once the transcription is completed, it can be easily downloaded in the formats TXT, SRT or WEBVTT. Transcribed files are available for download for one year.

Then follow the guide in Autotekst to add subtitles.