--- title: "Name the speakers in a meeting" description: "Listen to reliable voice samples, name the speakers, or finish processing first." updated: "2026-10-20" canonical: https://docs.recordstead.app/people/name-speakers --- # Name the speakers in a meeting When Recordstead has separated a meeting into voices but cannot name all of them, it asks for your review. Its row says **Waiting for names** while it waits for your first answer. The transcript is already saved. ## Open the window Click the **Review Speakers** notification, choose **Recording** ▸ **Review Speakers…** (⌘N), or click **Review Speakers** on Home. The window does not open by itself. **Review Speakers** above a meeting's transcript opens **Edit Speakers** instead, which saves with **Save Changes**. ## Put a name to each voice 1. Play a sample in each voice's row to hear who it is. Without a reliable sample, name the speaker only if you know who it is. 2. Enter the person's name in **Name**. With [saved-voice recognition](https://docs.recordstead.app/people/recognize-people) on, you can also pick a saved voice from the suggestions under the field. 3. Leave a row blank if you do not know who spoke. It remains an unidentified numbered speaker in the transcript. 4. Click **Save Names**. The names appear where those voices speak, and the meeting can continue to its summary. If processing was interrupted and no job is waiting, use **Generate** on the meeting’s Summary step. [View image: Speaker review window with separate voice rows, sample controls, Name fields and Save Names.]() Play each voice sample before giving that speaker a name. **Continue Processing**, closing the window or pressing Escape saves your draft and lets the meeting finish with its current speaker labels; name them later with **Edit Speakers…**. The review and its temporary voice data expire seven days after the review was created; the meeting and name editing remain. ## When two rows are the same person If Recordstead flags similar voices, [give the rows the same person](https://docs.recordstead.app/people/fix-speaker-voices) before saving. Each row remains a separate speech cluster. ## Change the names on a finished meeting Open the meeting's actions menu and choose **Edit Speakers…**. [Repairing speaker assignments](https://docs.recordstead.app/people/fix-speaker-voices) uses the same window. If its recording has been deleted, you can still rename a speaker, but cannot play samples. If it was moved to a mounted drive, samples load by themselves; for S3 or SFTP, click **Download** in the window first. A corrected name marks an existing summary as out of date; click **Update Summary** on its Summary tab to regenerate it. ## When it does not ask With [saved-voice recognition](https://docs.recordstead.app/people/recognize-people) enabled, a meeting whose speakers are all confidently recognized may finish without a review. Recognition is off by default. You can still edit those names afterwards.