Skip to main content
This feature is available on Pro plans and above. Turn on speaker diarization on a Lite plan and you’ll see an upgrade prompt.

​
How it works

Speaker diarization analyzes the full audio, splits it into who-said-what segments, and tags the transcript with labels like Speaker A and Speaker B. When this happens depends on how you record.
  • Audio file upload: applied as soon as the file is converted. Set the number of people in the [Number of Speakers] field at upload time for better results.
  • Offline recording: applied when the recording is converted after you stop.
  • Live recording: speakers aren’t split while you record. Instead, when you end the recording Tiro re-analyzes the full audio automatically, and the speaker tags appear in the transcript a moment later.

​
Turning on speaker diarization

1

Turn on speaker diarization in settings

Go to [Settings] → [Preferences] and turn on the [Speaker Diarization] toggle.
2

Record or upload a file

Record the way you normally do. For an audio file upload, set the number of people in the [Number of Speakers] field. Leave it on auto-detect if you don’t know.
3

Check the speaker tags in the transcript

Open the [Transcript] tab and you’ll see Speaker A, Speaker B tags on each spoken segment.

​
Renaming speaker tags to real names

You can rename the automatic tags to whatever you want, and the new names carry into document generation for more accurate documents.
  1. Click a speaker tag (Speaker A, etc.) in the transcript.
  2. Type a name directly, or pick one of your saved participants.
  3. Choose how far it applies: [Current section only], [From this section], or [Entire sections].
If diarization grouped two people into one segment, press Enter inside the paragraph to split the speaker yourself. You can also edit speaker tags in the mobile app.
You don’t have to rename the tags to use diarization. In that case your documents are written from the conversation alone.

​
Frequently asked questions

​
Why aren’t speakers split during live recording?

Live recording transcribes your audio in short chunks as it comes in, so it doesn’t split speakers while you record. Instead, when you end the recording Tiro re-analyzes the full audio automatically and adds the speaker tags. You can check them in the transcript right after recording ends.

​
Which plan includes speaker diarization?

Pro plans and above. Turn it on or off with the [Speaker Diarization] toggle under [Settings] → [Preferences].

​
A speaker is tagged incorrectly.

Click the speaker tag, rename it correctly, and choose how far the change applies. If one segment mixes two people, split it yourself with the Enter key. For an audio file upload, set the number of people in the [Number of Speakers] field next time. It improves accuracy.

​
Does renaming a speaker show up in the document?

Yes. Participants you tag yourself carry into document generation, so your meeting notes show real names instead of “Speaker A”.
Related pages: Manage conversation participants · Audio file upload · Offline recording