Optimization of Energy Consumption in Voice Assistants Through AI-Enabled Cache Implementation: Development and Evaluation of a Metric

Intelligent systems developed under the Internet of Things (IoT) paradigm offer solutions for various social and productive scenarios. Voice assistants (VAs), as part of IoT-based systems, facilitate task execution in a simple and automated manner, from entertainment to critical activities. Lithium...

Full description

Saved in:
Bibliographic Details
Main Authors: Alber Oswaldo Montoya Benitez, Álvaro Suárez Sarmiento, Elsa María Macías López, Jorge Herrera-Ramirez
Format: Article
Language:English
Published: MDPI AG 2025-01-01
Series:Technologies
Subjects:
Online Access:https://www.mdpi.com/2227-7080/13/1/19
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1832587422505893888
author Alber Oswaldo Montoya Benitez
Álvaro Suárez Sarmiento
Elsa María Macías López
Jorge Herrera-Ramirez
author_facet Alber Oswaldo Montoya Benitez
Álvaro Suárez Sarmiento
Elsa María Macías López
Jorge Herrera-Ramirez
author_sort Alber Oswaldo Montoya Benitez
collection DOAJ
description Intelligent systems developed under the Internet of Things (IoT) paradigm offer solutions for various social and productive scenarios. Voice assistants (VAs), as part of IoT-based systems, facilitate task execution in a simple and automated manner, from entertainment to critical activities. Lithium batteries often power these devices. However, their energy consumption can be high due to the need to remain in continuous listening mode and the time it takes to search for and deliver responses from the Internet. This work proposes the implementation of a VA through Artificial Intelligence (AI) training and using cache memory to minimize response time and reduce energy consumption. First, the difference in energy consumption between VAs in active and passive states is experimentally verified. Subsequently, a communication architecture and a model representing the behavior of VAs are presented, from which a metric is developed to evaluate the energy consumption of these devices. The cache-enabled prototype shows a reduction in response time and energy expenditure (comparing the results of cloud-based VA and cache-based VA), several times lower according to the developed metric, demonstrating the effectiveness of the proposed system. This development could be a viable solution for areas with limited power sources, low coverage, and mobility situations that affect internet connectivity.
format Article
id doaj-art-4d69b39b10ac4ef6ad482e6278c7cd78
institution Kabale University
issn 2227-7080
language English
publishDate 2025-01-01
publisher MDPI AG
record_format Article
series Technologies
spelling doaj-art-4d69b39b10ac4ef6ad482e6278c7cd782025-01-24T13:50:46ZengMDPI AGTechnologies2227-70802025-01-011311910.3390/technologies13010019Optimization of Energy Consumption in Voice Assistants Through AI-Enabled Cache Implementation: Development and Evaluation of a MetricAlber Oswaldo Montoya Benitez0Álvaro Suárez Sarmiento1Elsa María Macías López2Jorge Herrera-Ramirez3Faculty of Engineering, Instituto Tecnológico Metropolitano, Medellín 050013, ColombiaGrupo de Arquitectura y Concurrencia (GAC), University Institute of Cybernetics, Business, and Society, Universidad de Las Palmas de Gran Canaria, 35017 Las Palmas de Gran Canaria, SpainGrupo de Arquitectura y Concurrencia (GAC), University Institute of Cybernetics, Business, and Society, Universidad de Las Palmas de Gran Canaria, 35017 Las Palmas de Gran Canaria, SpainFaculty of Exact and Applied Sciences, Instituto Tecnológico Metropolitano, Medellín 050013, ColombiaIntelligent systems developed under the Internet of Things (IoT) paradigm offer solutions for various social and productive scenarios. Voice assistants (VAs), as part of IoT-based systems, facilitate task execution in a simple and automated manner, from entertainment to critical activities. Lithium batteries often power these devices. However, their energy consumption can be high due to the need to remain in continuous listening mode and the time it takes to search for and deliver responses from the Internet. This work proposes the implementation of a VA through Artificial Intelligence (AI) training and using cache memory to minimize response time and reduce energy consumption. First, the difference in energy consumption between VAs in active and passive states is experimentally verified. Subsequently, a communication architecture and a model representing the behavior of VAs are presented, from which a metric is developed to evaluate the energy consumption of these devices. The cache-enabled prototype shows a reduction in response time and energy expenditure (comparing the results of cloud-based VA and cache-based VA), several times lower according to the developed metric, demonstrating the effectiveness of the proposed system. This development could be a viable solution for areas with limited power sources, low coverage, and mobility situations that affect internet connectivity.https://www.mdpi.com/2227-7080/13/1/19voice assistantmetricartificial intelligenceenergy savingcacheintelligent systems
spellingShingle Alber Oswaldo Montoya Benitez
Álvaro Suárez Sarmiento
Elsa María Macías López
Jorge Herrera-Ramirez
Optimization of Energy Consumption in Voice Assistants Through AI-Enabled Cache Implementation: Development and Evaluation of a Metric
Technologies
voice assistant
metric
artificial intelligence
energy saving
cache
intelligent systems
title Optimization of Energy Consumption in Voice Assistants Through AI-Enabled Cache Implementation: Development and Evaluation of a Metric
title_full Optimization of Energy Consumption in Voice Assistants Through AI-Enabled Cache Implementation: Development and Evaluation of a Metric
title_fullStr Optimization of Energy Consumption in Voice Assistants Through AI-Enabled Cache Implementation: Development and Evaluation of a Metric
title_full_unstemmed Optimization of Energy Consumption in Voice Assistants Through AI-Enabled Cache Implementation: Development and Evaluation of a Metric
title_short Optimization of Energy Consumption in Voice Assistants Through AI-Enabled Cache Implementation: Development and Evaluation of a Metric
title_sort optimization of energy consumption in voice assistants through ai enabled cache implementation development and evaluation of a metric
topic voice assistant
metric
artificial intelligence
energy saving
cache
intelligent systems
url https://www.mdpi.com/2227-7080/13/1/19
work_keys_str_mv AT alberoswaldomontoyabenitez optimizationofenergyconsumptioninvoiceassistantsthroughaienabledcacheimplementationdevelopmentandevaluationofametric
AT alvarosuarezsarmiento optimizationofenergyconsumptioninvoiceassistantsthroughaienabledcacheimplementationdevelopmentandevaluationofametric
AT elsamariamaciaslopez optimizationofenergyconsumptioninvoiceassistantsthroughaienabledcacheimplementationdevelopmentandevaluationofametric
AT jorgeherreraramirez optimizationofenergyconsumptioninvoiceassistantsthroughaienabledcacheimplementationdevelopmentandevaluationofametric