Browse Prior Art Database

system method to detect, analyze and improve a caller's mood

IP.com Disclosure Number: IPCOM000231064D
Publication Date: 2013-Sep-25
Document File: 2 page(s) / 26K

Publishing Venue

The IP.com Prior Art Database

Abstract

Personal mood can impact the result and effectiveness of a phone conversation. Current technology normally uses pre-selected music or a wait time notification during the caller's waiting time. The drawbacks of the existing technology are that the music selection is not optional and repetitive, when there are extended wait times and the music is not desired. This can impact the caller's mood and lead to additional frustration or boringness for the caller. Scenario: A user calls the cable company and after the user speaks to the operator they transfer the user to technical support. There may be an initial hold time prior to being transferred which may involve an additional hold time of about 20 minutes. While user waits on hold user is obligated to listen to the hold music which user might not like and will leave a user irritated and unhappy. By the end of the 20 minutes the user is ready to get this call over with. User is now talking to technical support and instead of being relaxed and happy, user is frustrated, the user has wasted 20 minutes and basically an unhappy customer. This in turn makes conversation with technical support ineffective and frustrating for both parties. This may negatively impact the company's satisfaction indicators translating into loss revenue if customers are provided better customer service from a competitor. Implemention of this idea in call centers increases customer satisfaction and productivity. Happier customers means returning customers which results in a more productive and successful business.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 2

system method to detect, analyze and improve a caller's mood

The main idea is to use computer intelligence and analysis to detect, analyze and improve the mood of the caller and agent.


· Proposed is a mood recognition system where the system detects the user's mood based on the analysis of the user's voice volume, frequency/pitch, and tone, and suggests a selection of music or other audio programs (such as News, Weather , Sports..) to result into a calmer and happier caller.


· The mood detection system can also analyze the time of day and time of year to understand the emotional state of the caller. That is, if the caller is calling during rush hour, on a weekend, on a holiday...etc.


· The mood detection system can be based on various sensors to help to detect the user's mood such as temperature sensors to detect if the user is having a fever, smell sensor to see if the user is drunk, facial sensors to see if the user is angry or happy, etc .


· The mood detection system can also analyze the user's mood based on the user's historical activities as profiled in history records of user's last calls.


· The system routes the calls appropriately to improve the user's mood based on the mood detection of the caller and agent. Using this above user's mood analysis data, the system then recommends a list of options for the caller: that is, different types of music, weather news, sports...etc. Add perspective of detecting the agent's mood too. Once the mood is detected for the caller, implement one of the options to route the calls that are identified as potential "escalation" due to the mood detection (ex. how many times the clients got transferred), or route a special experienced agent group to handle bad mood clients, or changed the call waiting time to reduce further bad mood.


· The system routes the calls appropriately to improve the agent's mood using threshold based on the mood detection of the caller and agent. For example no agent would take more than 30% of the bad mood calls.

The novelty of this invention is:


· Using system intelligence and analytics to detect and analyze more accurately the users' mood based on multiple dynamic i...