System and method for searching content using acoustic characteristics
Content, such as communications, can be searched by their acoustic characteristics, among others. A method may include analyzing a plurality of content for at least one of a language characteristic or a non-word symbol within a text of the plurality of content, and an acoustic characteristic, searching the plurality of content using a search string comprising at least one of a language characteristic, or a non-word symbol, and an acoustic characteristic, and presenting a subset of the plurality of content ranked according to a match with the search string. In some cases, a category label is determined for the content and is also used in the search string.
1. A method, comprising:
analyzing acoustic and language characteristics of a plurality of communications between customers and agents to identify communications that include acoustic agitation or dissatisfaction expressed by a customer and to identify a response by an agent in the identified communications;
determining at least one category label to associate with one or more of the plurality of communications;
determining relevant communications of the plurality of communications relating to a specific agent based on a criteria that is at least one of a category label or a presence of acoustic agitation or dissatisfaction expressed by one or more customers, wherein determining the relevant communications further includes restricting the plurality of communications to those with a specified amount of time between at least two specified events;
analyzing the relevant communications to determine a quality score for the specific agent relating to a responsiveness of the specific agent to the acoustic agitation or the dissatisfaction of the one or more customers;
providing the plurality of communications, the at least one category label associated with one or more of the plurality of communications, and the quality score in a social feed;
receiving one or more comments in the social feed on one or more of the plurality of communications or the quality score;
sorting the one or more comments by a seniority of commenter to obtain at least one high interest comment; and
presenting the at least one high interest comment to a manager of an agent.
2. The method of claim 1 , wherein at least one of the at least two specified events is an acoustic event.
3. The method of claim 2 , wherein the acoustic event is at least one of a silence, an overtalk, a change in tempo, a gain in a volume of words, or a time lag between words.
4. The method of claim 1 , wherein at least one of the at least two specified events is an alert.
5. The method of claim 4 , wherein the alert is at least one of an acoustic alert, a conversational alert, or a call flow alert.
6. The method of claim 1 , wherein at least one of the at least two specified events is an escalation attempt, a use of profanity, a sales opportunity, or a compliance violation.
7. The method of claim 1 , wherein at least one of the at least two specified events is an occurrence of an item in a list for the category label.
8. The method of claim 7 , wherein the item in the list for the category label comprises a language pattern, a keyword, a phrase, a non-word symbol, a behavior, or an emotion.
9. The method of claim 1 , wherein the specified amount of time is selectively variable.
10. The method of claim 1 , wherein the specified amount of time is a time period before one of the at least two specified events.
11. The method of claim 1 , wherein the specified amount of time is a time period after one of the at least two specified events.
12. The method of claim 1 , wherein the at least one category label comprises a combination of two or more other category labels.
13. The method of claim 12 , wherein the combination of the two or more other category labels comprises a combination of two or more of: a behavior, a reason, a procedure, a competitor, a dissatisfaction, an agitation, a satisfaction, an escalation, an empathy, a repeat contact, a transferred call, and a politeness.
14. The method of claim 1 , wherein the plurality of communications are voice communications.
15. The method of claim 1 , further comprising filtering the plurality of communications using metadata, wherein the metadata is at least one of a speaker name, a speaker type, a communication type, an agent data, an agent grouping, a call handling location, a time and date of a call, a department, a skill or queue, a pertinent IVR path data, another category, or a call length.
16. The method of claim 1 , wherein the plurality of communications are voice communications and the method further comprises converting the voice communications to text using a computer-based speech recognition module.
17. The method of claim 1 , wherein sorting further comprises sorting the one or more comments by at least one of a frequency of comment or an extent of agitation expressed by one or more customers to obtain at least one high interest comment.