Topics are a collection of phrases defined globally for the SpeechMiner system, and included in one or more programs as required. When a topic is included in a program, SpeechMiner searches for all of the phrases included in the topic's definition in all the interactions associated with the program. When one of the phrase is found, SpeechMiner identifies it as a linguistic event and registers the topic as found at a specific time during the interaction.
When you create a topic, include a wide range of numerous phrases that can be associated with the same business issue.
Although a topic can contain one phrase, the more phrases you add the more useful the topic. Also, more than one topic can contain the same phrase.
Define initial Phrases and Keywords

Each phrase in the external file should be on a separate line.
The last phrase in the list of phrases appears in the field.
When a new phrase is added to a SMART topic, the system verifies whether or not the phrase is long enough to be accurately identified in the interaction. If the phrase is not long enough, a Too Short Terms warning message appears and a red icon (
) is added to the specific phrase in the Text list.
Your selection appears in the field above the list and the New button becomes an Update button.
| Although you can leave the short phrase in the list as is, the system may not recognize it in an interaction. |
If the system finds a match, you can use the existing phrase instead of creating a new one, or modify your entry to create a new phrase.
If a phrase is a duplicate of an existing phrase, a red M flag is inserted beside the phrase and the duplicate phrase is displayed with blue text as follows:
.
For additional information, refer to the red M flag tool tip.
| Genesys recommends to check all new phrases at once, since this verification process requires significant system resources. |
By default, each topic phrase is assigned the same strictness value assigned to the topic itself.
The strictness given to a phrase overrides the topic's strictness. When a phrase is given a strictness level, the level appears beside the phrase in the list
.
The level of strictness defines how confident SpeechMiner must be to recognize a topic phrase. For example, if you select Medium (65), the topic phrase must be identified with confidence levels of at least 65 in order to be identified by SpeechMiner. For example, if I have the phrase with a confidence of 60 and the topic was configured with a strictness of 70, it will be marked as not found. If the strictness is 50 it will be marked as found.
The Term Strictness window opens.
Repeat any of the steps in this procedure until you have a collection of phrases that represent the business issue you are interested in.
Assign Topic Keywords and Escapes