Manual enrollment by interaction import

Usually the interactions used to train voiceprint models are on a Recorder in the suite. However, you can also import audio from an external source onto the recording platform.

Audio import converts the audio file to the standard recording format of the Recorder. Appropriate metadata is used to create an interaction in the database. After audio is ingested, all Recorder operations such as enrollment, archive, and search and replay can be performed on the interaction.

Importing audio for manual enrollment can temporarily affect automatic enrollment in the following cases:

  • If the Recorder Analytics Rule used for automatic enrollment uses the Minimum Days between Calls setting, an import can block voiceprint enhancement by automatic enrollment until the number of days configured in the rule elapse.

  • The interaction start time for imported audio is the time that the import occurs. If the import occurs during a period when capture occurs for an interaction, the captured interaction would not be used for automatic enrollment. To prevent the duplication of interaction usage in N+N environments, automatic enrollment rejects an interaction for the same person with a capture time that overlaps another interaction.