Skip to content

Topics and keywords

MiaRec allows you to record and analyze 100% of calls automatically. It does the following actions:

  • Identifies each defined keyword/phrase that the participants use in a conversation
  • Categorizes calls by topics
  • Calculates the aggregated call score
  • Displays call volume trends by topic over a period of time
  • Searches calls by topic, keyword or score

This capability negates the need to spend resources on manual analysis of random recordings and provides complete coverage of voice interactions in the contact center.

Keyword spotting

Keyword spotting, a subset of speech analytics, is the ability of a monitoring system to recognize predefined words and phrases in interactions.

For example, you are interested in knowing when customers use the word "frustrated" or other words during an interaction with one of your agents. You define the keywords in the MiaRec application and then put it into operation.

Examples of keywords / phrases:

  • "frustrated"
  • "upset"
  • "cancel my account"
  • "angry"
  • "you're not listening"

MiaRec identifies who spoken the spotted keywords, agent or customer. Some keywords may have different value or even meaning depending of who, agent or customer, speaks them. For example, phrases like "thank you so much", "excellent", "fabulous" are more valued if they are spoken by customer rather than by agent, who is trained to be polite during a call.

MiaRec shows the spotted keywords above the transcription as well as highlights them within a transcription.

Keyword spotting

Topic extraction

A topic is a set of similar or related keywords that fall into the same category. For example, a topic "Repeated calls" may consist of phrases like "called before", "called twice", "called last week", "never heard back" etc.

Examples of topics:

  • Upset customer
  • Account cancellation
  • Repeated calls

MiaRec shows the extracted topics in call details.

Topic extraction

Info

Topic extraction is highly effective when used together with the Sentiment score feature.