The online version of Microsoft Word (M365) includes a transcribing feature that allows you to easily and quickly convert a video or audio recording of an interview to text, for example. The service is available to both staff and students, and dictation/transcribing can be used for 300 minutes per month.

It is not allowed to transcribe audio recordings containing human voices unless these recordings are already publicly available online. If the material contains personal data or confidential data, please contact Digital Services data security experts at HelpJYU-service or via email: tietoturva@jyu.fi

This service can not create video captions, because timestamps are not compatible with WebVTT/SRT-formats that are used in video captions.

Create a new blank document in the online version of Word / Microsoft 365 (Uusi tyhjä asiakirja).

Open the Home tab (Aloitus) of the toolbar and click the down arrow next to the "Dictate" (Sanele) icon. Select "Transcribe" (Litteroi) from the drop-down menu.

The transcribe function opens on the right side of the window. Now you can choose to record a new audio right away or select an existing audio or video file. The recordings can be either .wav, .m4a or .mp3 audio files or .mp4 video files. This guide covers transcribing of a ready-made recording.

Click the Upload audio button. Select the desired audio or video file in the file browser window and click Open.

Depending on the length of the recording, it may take some time to transcribe it. Leave the browser window open until process is complete:

After transcribing is complete, you can preview the text before adding it to the Word document. Incorrectly recognized sections can be edited by hovering over the text chapter and clicking the pencil icon (Edit transcript selection). You can also listen to the recording by pressing Play button.

When the edits are complete, you can add all the text to the Word document by clicking "Add to document". You can choose from four different text formatting options:

- Just text = plain text without any additions
- With speakers = Text and information about who is currently speaking.
- With timestamps = Text and timestamps
- With speakers and timestamps: Text, timestamps and speaker information.