Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

System and Method for Improving Time Utilization of People who Must Make Phone Calls during Working Hours

IP.com Disclosure Number: IPCOM000246172D
Publication Date: 2016-May-13
Document File: 5 page(s) / 67K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a method and system for improving the time utilization of a caller/employee who is on hold waiting for support service during work hours. Using voice recognition and natural language processing technologies to identify active conversation, the system implements a mediator function, which places the caller/employee in a temporary environment that allows the caller/employee to both remain attentive to the call, listening for a response, and be productive while working.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 34% of the total text.

Page 01 of 5

System and Method for Improving Time Utilization of People who Must Make Phone Calls during Working Hours

Many organizations provide customer service phone lines that are functional only during business hours (e.g., 8:00am-5:00pm, weekdays). If an employee (or self-employed person) makes a call to a customer service line during

work hours and must wait for assistance (i.e., on hold) for an extended period, then this is not only bothersome to the caller, but it results in a loss of productivity during working hours.

A method and system are needed that can improve the caller's time utilization, as well as overall human experience,

while waiting for assistance from a representative.

Substantial developments in the fields of speech recognition and natural language processing enable the disclosed solution.

The novel contribution is a method and system that installs a call mediator . The main innovation is the automatic detection of the human agent and re-connection of the caller with neither the caller's continuous attention nor the human agent's cooperation. The mediator places the caller/employee in a temporary environment that allows the caller/employee to both remain attentive to the call, listening for a response, and be productive while working. This can also reduce the feelings of annoyance the caller/employee might have while waiting for a response.

The mediator is implemented through a mobile device application or a physical device attached to traditional phone sets. The system understands the incoming contents while the caller is waiting. If the caller is receiving repeated commercial messages, distracting music, etc. then the system can replace that audio content with content that is preferable to the caller. The system recognizes when a human voice (i.e., the representative) begins speaking, so the caller can take over the line. The mediator also engages in a brief conversation with the human agent until the human caller takes over.

Learning

Recognition of a human voice, the speaker identity, and the text spoken is now feasible with a high level of accuracy with current technologies. It is possible to turn voicemail messages, for example, into text without special training for particular speakers. A higher accuracy level is achievable with specific training for the speaker. The speaker identification further allows the system to recognize the speaker name and identifier (ID) from a large database of speakers acquired by the system through training. Consequently, the novel system can be trained for specific often-called service centers (e.g., IRS, Social Security Administration, U.S. Citizenship & Immigration Services, etc.). The specific training can be based on multiple calls and hold periods from the customer-service provider, which is recorded and labeled as "hold" and "agent".

1


Page 02 of 5

A computer system is then trained to classify the audio signal to one of these classes by combining both textual content

and spea...