Contact Center Audio Analysis During Non-Audio Modes

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Contact center systems lack effective mechanisms to analyze and utilize audio signals captured by microphones during non-audio communication modes, such as when agents are not actively conversing with individuals, leading to missed opportunities for data capture and analysis.

Innovation Solution

Implementing a system that identifies and analyzes utterances spoken by contact center agents during non-audio modes, generating session metadata, and storing it in contact session records, which can include commands, profanity flags, and sentiment analysis, to facilitate various applications and actions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If the microphone is utilized to capture sounds during non-audio communication modes, then audio data is captured and analyzed for performance evaluation and quality assurance, but system complexity increases due to the need for audio analysis processing during inactive periods

Engineering Contradiction:
Improveaudio data captureVSAvoidsystem complexity
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The system performs preliminary classification of audio signals during non-audio modes to identify utterances before full analysis. This preliminary action filters out irrelevant sounds and prepares only significant audio segments for detailed processing, reducing overall system complexity while maintaining comprehensive audio data capture capability

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system extracts only the necessary audio segments containing utterances during non-audio communication modes, separating them from the continuous audio stream. This extraction approach captures valuable audio information while avoiding the complexity of processing entire audio streams, focusing resources only on relevant data

Inventive Principle:
Principle #2Taking out (Extraction)

2Loss of information

If audio analysis is performed during non-audio modes, then valuable metadata is generated for performance evaluation, but processing time and computational resources are consumed during periods when the agent is not actively communicating

Engineering Contradiction:
Improvemetadata generationVSAvoidprocessing time
Core Design Contradiction:
Loss of informationVSLoss of time

Solution Approach 1:

The system implements periodic audio analysis during non-audio modes rather than continuous processing. It analyzes audio signals at scheduled intervals or when specific conditions are met (such as detecting potential utterances), generating necessary metadata while minimizing unnecessary processing time and computational resource consumption during agent idle periods

Inventive Principle:
Principle #19Periodic action

3Loss of information

If the system distinguishes between audio and non-audio modes, then audio analysis is applied selectively to improve relevance, but the difficulty of detecting and measuring communication mode increases

Engineering Contradiction:
Improveaudio analysis relevanceVSAvoidcommunication mode detection
Core Design Contradiction:
Loss of informationVSDifficulty of detecting and measuring

Solution Approach 1:

The system uses feedback from the communication platform to automatically determine whether the agent is in audio or non-audio mode. By monitoring the active communication channel and agent status information, the system selectively applies audio analysis only during non-audio modes, improving analysis relevance while avoiding the complexity of implementing complex mode detection algorithms

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS10218841B2Microphone monitoring and analytics
Publication Date: 2019.02.26 AVAYA INC
  • US10218841B2 patent drawing
  • US10218841B2 patent drawing
  • US10218841B2 patent drawing

AI summary

Microphone monitoring and analytics are provided. An initiation of a new contact session is determined that includes a communication interaction between a contact center agent (CCA) having a microphone and a contacting individual (CI). Prior to an end of the new contact session, a non-audio mode wherein audio signals received by the microphone are not communicated to the CI is determined. Audio signals received via the microphone during the non-audio mode are analyzed. An utterance spoken by the CCA is identified. Contact session metadata is generated based on the utterance. A contact session record that includes the contact session metadata is generated. The contact session record is stored.