Skip to main content

🔖 Speaker Labeling

Label your speakers so the AI has context about who is who in the recording.

Blaine Bolus avatar
Written by Blaine Bolus
Updated over a year ago

Labeling your speakers is necessary to give the AI context. Follow these simple steps for magic:

Step 1: Click into your first episode you’ve uploaded

Step 2: Click the “Speaker A” “Speaker B” etc at the top of the transcript
​


​Step 3: Click “add new speaker” and designate role as a guest or a host
​

Step 4: Label ALL speakers, even if you need to label someone twice
​

Step 5: Click the AI CONTENT tab to see your generated results
​
​Here's what it looks like:
​


​

Once you've labeled the speakers, our AI will whip up improved content. This includes titles, speaker bios, introductions, timestamped overviews, topics & bullets, keywords, quote hooks, LinkedIn posts, newsletters, reel scripts, tweet threads + more. Plus, you can create better custom content with our prompt interface! 🧙
​
​What if speakers are incorrect?

AI isn’t always perfect here, and we’re working to make this as accurate as possible. in the meantime if you run into a corner case, here’s what to do:

  • label the speakers to align with their majority of content

  • if the same person is split into 2 speakers, label them both the same name and merge them together

  • if wrong in the transcript, click on the speaker name and select from the dropdown

  • if a speaker is omitted, try to re-upload using the “expected speakers” parameter or leave blank (opposite of what you did first time around)

Edit button

This let’s you make minor edits to the transcript. You can also change the speaker name if it was mislabeled by AI

Check out this loom video for more:

Did this answer your question?