Add a voice sample to a profile
Record or upload a short sample so the profile gets a voiceprint.
Where to find it
Voice › Voice profiles
Open this screen
Steps
- In the sidebar, open Voice › Voice profiles.
- Press Record a sample on the profile (Add a sample if it already has a voiceprint).
- Press Record, allow the microphone, and let only this person speak for about 20 to 30 seconds in a quiet place. It stops by itself after 30 seconds, or press Stop.
- Or choose a file next to or upload a file.
- When it says Sample ready to send, press Save voiceprint.
Settings on this screen
| Setting | What it does | Default |
|---|---|---|
| Record | Records from the browser's microphone, up to 30 seconds. | — |
| or upload a file | Uses an audio file instead. | — |
| Save voiceprint | Sends the sample to make the voiceprint. | — |
Troubleshooting
"The microphone could not be opened."
Allow the microphone for this site in the browser, then press Record again.
"Voice recognition needs setup."
No voice-recognition key is configured for the service, so nothing can be enrolled. Ask the service operator.
"The voice sample could not be processed."
Record again with only one person speaking, in a quiet place.
Tips
- The sample is not kept here; only the encrypted voiceprint is.
Good to know
- Only workspace managers can record samples.
- The profile must have an active consent record.
Related how-tos
- Voice profiles Register known voices, with consent, so call and meeting transcripts show their name instead of Person A.
- Add a voice profile with consent Register a client's or staff member's voice profile and record their consent.
- Clone your own voice Record your own voice so your AI agent can speak on calls in it.