> ## Documentation Index > Fetch the complete documentation index at: https://docs.granola.ai/llms.txt > Use this file to discover all available pages before exploring further. # Speaker attribution > Granola can label who said what in your meeting transcript using each participant's display name. Set it up for Google Meet or Zoom. Speaker attribution labels who said what in your transcript using each participant's display name, instead of just **Me** and **Them**. When Granola can identify someone speaking, their name appears above their parts of the conversation. ## Set up speaker attribution Speaker attribution is set up per meeting platform. Choose yours below for setup instructions. Available on macOS and Windows through the Granola browser extension. Available on macOS through **Settings > Preferences**. ## Use speaker attribution in a meeting Once set up, there's nothing extra to do in the meeting itself. Start your call and start transcription in Granola as normal, then open the transcript panel to see live transcription with speaker names. Speaker attribution only applies while Granola is actively transcribing — it does not add names to older transcripts. It also does not start transcription by itself: Granola only transcribes when you open a meeting note or start a New Note. Learn more in [How transcription works](./transcription). ## What speaker attribution can and cannot identify * **People joining the call individually**: Granola uses the display name your meeting platform shows for each participant. * **People sharing a meeting-room device**: If several people speak through the same shared device, Granola cannot distinguish the individuals using it. * **Audio outside the call**: Audio playing from elsewhere on your computer may be transcribed without a matching speaker name. * **Overlapping speech**: If several people speak at once, Granola may not be able to assign every part of the conversation to the correct person. Where speaker attribution isn't available or set up, transcripts show **Me** and **Them**, corresponding to your microphone input and your system audio. Even then, our AI usually does a good job of inferring speakers from contextual clues in the transcript, and you can correct a name directly in your notes or ask Granola Chat to update it. ## Privacy and in-meeting notice Speaker attribution does not record audio or video — Granola captures audio for transcription as described in [How transcription works](./transcription). Speaker attribution is not an alternative feature for notifying participants that you are using Granola. It is merely attributing the identity of speakers in your transcript. If you or your organization want to let participants know Granola is being used, see [our transparency features](../consent-security-privacy/transparency-solutions/introduction).