How to automatically detect speakers in an audio or video recording

Interviews, podcasts, panel discussions, meetings, Gaston automatically figures out how many people are speaking in a recording and labels every sentence with who said it, no manual tagging required.

1. Add your recording as usual

Add the file the same way you would any other, by uploading it or pasting a link. Speaker detection runs automatically as part of transcription, there's nothing extra to turn on.

2. Tell Gaston the number of speakers, if you know it

Speaker detection works out of the box, but it's most accurate when you tell it exactly how many people are talking. If you know the number in advance, set it in the speaker detection settings before transcription finishes.

3. Read the labeled transcript

Once transcription is done, open the file in the player. Each sentence is grouped by speaker, so you can immediately see who said what without listening to the whole recording.

4. Search or export by speaker

Speaker labels carry through to search and export too, so you can pull out everything one person said, or generate a transcript that clearly separates each voice.

Now that every sentence knows who said it, searching a multi-speaker recording gets a lot more useful, see the search guide below.