Home > Overview > Events: Topics and Non-Linguistic Data

Events: Topics and Non-Linguistic Data

Events are specific types of call content that are identified by SpeechMiner in calls. SpeechMiner can identify linguistic Events - that is, Topics - and certain types of non-linguistic Events.

Topics

A Topic represents a specific business idea that you want to detect in analyzed calls. For example, if you want to identify calls in which the customer wishes to cancel an account, you could create a Topic named Cancelation.

Once a Topic is defined, you populate it with terms - phrases that correlate with the idea the Topic was meant to capture. For example, you could add the following terms to the Cancelation Topic:

Topics can contain one or more terms, but at least 15 terms are usually necessary to define a useful Topic. A Topic can even contain hundreds of terms, if necessary. All of the terms should be concrete ways of identifying the idea represented by the Topic in a conversation.

Topics are defined globally for your entire SpeechMiner system. In order to use them, they must be linked to Programs. When a Topic is linked to a Program, SpeechMiner searches for the terms of the Topic in calls that belong to the Program. If it finds one of the terms, it registers the Topic as found in that location in the call.

Each Topic Event that is identified by SpeechMiner has a start time, an end time, a name, and a type. For example, if a Topic called "Loan Offer" is identified by the term "can offer you a loan," the start time is when the agent began saying "can," the end time is when the agent finished saying "loan," the name is "Loan Offer," and the type of Event is "Topic."

Non-Linguistic Events

Non-Linguistic Events are sounds other than speech. SpeechMiner automatically identifies non-linguistic Events in calls, as follows: 

In addition, when you configure a Program, you can choose whether SpeechMiner should also identify agitation - non-verbal expressions of frustration and anger, such as deep sighs, grunts or rapid changes in pitch.

Each non-linguistic Event that is identified by SpeechMiner has a start time, an end time, and a type. For example, if SpeechMiner identifies silence in a call, this is a non-linguistic Event whose start time is the beginning of the silent period, whose end time is the end of the silent period, and whose type is "Silence."

 

See also

Programs
Call Classification: Categories
Language Support
Overview of the Interface
Typical Workflows