Call analysis model updates

Each country has a unique combination of carrier tones and spoken carrier messages that are played to callers, for example, carrier tones received by Outbound Dialer. The accuracy of detecting these audio events is continuously improved, over time, across the world approximately on a monthly basis. Spoken carrier messages, such as SIT-related, busy, and machine, can be detected with high accuracy by using audioprints, also known as fingerprints. These audioprints are delivered by using the model update process. If carrier messages or tones are mis-detected during Outbound dialing, contact Customer Care. 

For premises, all LDM Edges across the organization must be upgraded to the latest version to benefit from the model updates in the latest version. Otherwise, the model associated with the oldest Edge version in the organization is used for all Edges, regardless of the Edge version. 

For cloud media CDM Edges, an Edge update is not required, as the model update is delivered by using cloud media services. 

For a list of supported call analysis regions where these models are used, see Supported call analysis regions.