When diio receives audio with more than one voice, it identifies each one and labels them "Voice 1", "Voice 2", and so on. It needs you to tell it who each one belongs to, because commitments, mood, and performance are calculated per person. An incorrect assignment distorts the entire analysis of who said what.
This applies to meetings, including the in-person ones you send to the diio chat on WhatsApp. Voices can't be reassigned on calls.
During the WhatsApp upload
This is the most convenient moment, because diio asks you directly in the chat:
diio shows you a clip of each detected voice.
For each one, select the participant's name or email.
Confirm. If there are more than two people, the process repeats for each voice.
Later, from the detail view
If you didn't assign them during the upload, or you want to correct an assignment in a meeting:
Open the conversation in "All conversations".
On the Participants card, click "Edit".
Reassign each voice to the right participant.
Save. diio recalculates the analysis with the updated assignment.
The "Edit" button on the Participants card only exists for meetings. Voices can't be reassigned on calls.
What happens if you don't assign the voices
The analysis is generated anyway, but commitments and metrics are attributed to "Voice 1" or "Voice 2" instead of to the actual participants. This makes follow-up harder and can cause errors when syncing to the CRM.
A different case: unscheduled meetings
If you used the Chrome plugin to record a meeting that wasn't on your calendar, the assignment doesn't happen through WhatsApp but from "Pending recordings" in the plugin settings. There you enter the name and email for each voice before diio processes the conversation.