Configuring Text Elements

A Text element specifies that a phrase containing one or more words must be found (or must not be found) by SpeechMiner in the transcript of a call in order for the call to be assigned to the Category. The phrase must be found exactly as it is entered in the Text element definition to meet the condition, except that it is not case sensitive. For example, "i want to repay my loan" and "I want to Repay My loan" are treated as identical phrases, but "I wanted to repay my loan" and "I want you to repay my loan" are not identified as matching them.  

When you add a Text element to an expression (see Inserting Elements into an Expression), the fields required to select and configure the element are added to the right panel below the Type field. You can use these fields to specify the phrase, its required number of occurrences in the call, its strictness level, and its timing constraints.  

To add a Text element:

  1. Under Text, enter the phrase. The phrase should consist of a sequence of one or more whole words. The words should be separated by single spaces, and no punctuation or other symbols should be included in the sequence. The text is not case-sensitive. 

  2. Modify the following settings as necessary:

    Setting Description 
    Not found Select this option if you want the condition to include all calls except those in which the phrase was found. (For additional information, see Configuring Category Conditions.)
    At least __ Occurrences Specify the minimum number of times the phrase must have been found in the call in order for the condition to be met.
    At least __ Confidence Select the minimum level of confidence for the recognition of the phrase; if the Confidence level does not meet this minimum, SpeechMiner proceeds as if the phrase was not found.

    From

    Define the earliest position in the call audio at which the phrase may be found, in seconds, and then select one of the following types of reference points:

    • Secs Abs: The number of seconds is defined relative to the beginning of the call
    • Secs Abs from End: The number of seconds is defined relative to the end of the call (i.e., the phrase must be found no earlier in the call than this number of seconds from the end of the call)
    • Secs Rel to Start of Prev: The number of seconds is defined relative to the beginning of the previous item in the expression; this option will only work if the operator connecting the two elements is Conditional AND (->). Otherwise, selecting this is the same as selecting Secs Abs.
    • Secs Rel to End of Prev: The number of seconds is defined relative to the end of the parent item; this option will only work if the operator connecting the two elements is Conditional AND (->). Otherwise, selecting this is the same as selecting Secs Abs from End
    To Define the latest position in the call audio at which the phrase may be found, in seconds, and then select one of the types of reference points explained above.
    Speaker Type  Select a speaker type if the phrase must be found when the selected type of speaker is speaking, or ANY to include all speaker types. 

 

See also

Configuring Metadata Elements
Configuring Non-Linguistic Event Elements
Configuring Program Elements
Configuring Topic Elements
Configuring Category Elements
Configuring Speaker-Type Elements