Voice Recognition Call Termination for Energy Reduction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing systems for terminating voice calls require user intervention or predefined commands, leading to issues like energy waste and unnecessary communication band usage when calls are forgotten.

Innovation Solution

A system utilizing artificial intelligence and voice recognition technology to automatically terminate calls by detecting predefined keywords and learning new end-of-conversation words, applicable across various languages and devices.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of energy

If automatic call termination system is implemented, then energy waste and communication band usage are reduced, but device complexity increases

Engineering Contradiction:
Improveenergy wasteVSAvoiddevice complexity
Core Design Contradiction:
Loss of energyVSDevice complexity

Solution Approach 1:

The system enables the device to automatically terminate calls without user intervention by detecting keywords and analyzing conversation patterns. The device serves itself by monitoring its own call state, recognizing end-of-conversation keywords, and autonomously terminating the call, thereby eliminating energy waste from forgotten calls while managing complexity through automated self-management.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system continuously monitors the voice conversation for keywords indicating call end, checks conversation continuity, and provides feedback to the call termination control. This closed-loop feedback mechanism allows the system to accurately determine when to terminate calls automatically, reducing energy waste while maintaining controlled complexity through systematic monitoring and response.

Inventive Principle:
Principle #23Feedback

2Measurement precision

If voice recognition system analyzes conversation continuously, then call termination accuracy improves, but energy consumption increases

Engineering Contradiction:
Improvecall termination accuracyVSAvoidenergy consumption
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

Instead of continuous analysis, the system periodically checks the conversation for keywords at intervals. The voice recognition system monitors the call flow and performs keyword detection at specific points during the conversation, balancing termination accuracy with reduced energy consumption by avoiding constant analysis while still capturing end-of-conversation indicators effectively.

Inventive Principle:
Principle #19Periodic action

Solution Approach 2:

The system pre-loads and stores a dictionary of end-of-conversation keywords in the device memory before calls occur. This preliminary preparation allows the voice recognition system to efficiently match spoken words against predefined keywords during calls without requiring complex real-time language processing, thereby improving termination accuracy while minimizing energy consumption during actual call handling.

Inventive Principle:
Principle #10Preliminary action

3Device complexity

If predefined keywords are used for termination, then system simplicity is maintained, but adaptability to different languages and contexts is reduced

Engineering Contradiction:
Improvesystem simplicityVSAvoidlanguage adaptability
Core Design Contradiction:
Device complexityVSAdaptability or versatility

Solution Approach 1:

The system employs a multi-language keyword dictionary that stores end-of-conversation keywords in multiple languages. The voice recognition system can selectively activate appropriate language dictionaries based on the call context or user preferences, enabling the simple keyword-based termination system to adapt to different languages and cultural contexts without requiring complex language processing infrastructure.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system dynamically adjusts its keyword matching behavior based on conversation context. It monitors conversation flow, speaker turns, and temporal patterns to determine when detected keywords genuinely indicate call end versus when they are merely mentioned during ongoing discussion. This dynamic contextual analysis enhances adaptability to different conversation styles and languages while maintaining the simplicity of keyword-based detection.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP3422343B1System and method for automatically terminating a voice call
Publication Date: 2020.07.29 VESTEL ELEKTRONIK SANAYI & TICARET ANONIM SIRKETI
  • EP3422343B1 patent drawingFigure 1
  • EP3422343B1 patent drawingFigure 2

AI summary

The present invention refers to a system and a method for automatically terminating a voice call. The system for automatically terminating the voice call preferably comprises a memory unit (5) in a device configured to store a set of defined keywords (6) indicating the end of the voice conversation (7), and a voice recognition system (4) provided in the device and configured for recognizing a voice conversation (7) between a caller and a receiver, wherein the voice recognition system (4) is configured to detect at least one keyword of the set of defined keywords (6) in the voice conversation (7) of the voice call, wherein the voice recognition system is further configured to check whether the voice conversation (7) is continued or not for a predefined time after detecting at least one keyword, and wherein the voice recognition system is configured to terminate the call after detecting at least one of the defined keywords (6) if the voice conversation (7) is not continued.