Creating Voice Signatures
First you need to create signatures under your profile settings. It is best to create several recordings, possibly with different headsets or microphones, under different conditions. When you go to the User -> Profile -> Voice Signature for the very first time you will see this screen:
You then press the red Record button and you will this next dialog and the recording will start (On subsequent visits to this page you can start adding a new audio by clicking the + button in the top-right corner.)
You will see the suggested text to read but you do not need to follow it 100%. Once done click the stop button. You then have option to Restart, or Continue.
Here you can listen to the audio, give a name to the audio clip and add some optional tags. Once done you click save.
It takes a few seconds for the voice signature to be computed. Initially, the duration will not be shown, but after a while you can refresh the page.
If you see the duration then it means that the signature has been successfully computed.
Using Voice Signatures
In order to tell the system to use the Voice Signatures you need to specify one or more Expected Speakers on the Project->Settings->Transcription page:
Then you can run transcription, and once done and you click on the Transcript and show the details you will see the recognized speaker under Participants list - you will see the avatar, full name, and an icon indicating that voice signature was used to identify that speaker.
Managing Voice Signatures
If you are an Admin User then you will have access to a page that allows you to manage all the Speakers (with and without Voice Signatures). This page is under Profile settings:
