Feature Request: Global Speaker Renaming in Diarised Fulltext Transcriptions

[erklu]

Problem

In diarised fulltext transcriptions, speakers are automatically labeled (e.g. Speaker00, Speaker01).
Currently, renaming a speaker can only be done per block in the editor.

For recordings with many speaker turns (interviews, panels, hearings, qualitative research material), this is time-consuming and increases the risk of inconsistent speaker naming.

A common workflow is to identify who a diarised speaker actually is, and then apply that name consistently across the entire transcription.

Proposed Solution

Add a separate speaker panel, suggested placement together with other metadata beneath the media player with support for global speaker renaming in diarised fulltext transcriptions, allowing a user to rename a speaker once and have the change applied to all blocks that reference that speaker.

Expected Behaviour

• A speaker can be renamed in one action
• The new name is applied to all existing blocks using that speaker ID
• The change affects speaker metadata, not the transcribed text itself
• Renaming is local to the current transcription
• Internal speaker IDs (e.g. Speaker00) may be preserved internally for traceability

UX / Interaction (implementation-agnostic)

Possible interaction patterns include:

• A speaker list/panel showing all detected speakers with editable names
• A context action on a speaker label (e.g. “Rename speaker”)
• A simple dialog:
  • Select speaker
  • Enter new name
  • Apply (to all blocks)

Exact UI design is left to implementation.

Export Behaviour

• Renamed speaker labels should be reflected in fulltext exports where speaker information is included
• No automatic speaker merging or re-diarisation is implied

User Value

• Significant time savings when working with diarised fulltext material
• Reduced risk of inconsistent speaker naming
• Well aligned with qualitative research and interview-based workflows

Priority

Medium–High