STUDY OF ALGORITHMS APPLIED TO VOICE SIGNAL ANALYSIS: RECOGNITION OF VOICE PATTERNS USING ARTIFICIAL INTELLIGENCE

Authors

  • Amivaldo Batista dos Santos Instituto Federal Goiano
  • Solange da Silva Pontifícia Universidade Católica de Goiás
  • Pedro José Abrão Instituto Federal de Educação, Ciência e Tecnologia de Goiás
  • Luana Machado dos Santos Centro Universitário Araguaia https://orcid.org/0000-0002-1640-6262

Keywords:

Pattern recognition. Voice analysis. MFCC. AI-Mel Frequency.

Abstract

This study refers to the perspective of voice signal standard recognition using Artificial Intelligence (AI) through Artificial Neural Networks techniques (ANN), using the Mel Frequency Cepstral Coefficients (MFCCs), that extracts voice signal characteristics. The purpose of this paper was recognizing voice signal patterns from the Fire Department of the State of Goiás database. The literature review allowed us to analyse algorithms that analyse voice signals. Posteriorly, experiments with the same algorithms were done in order to find a pattern of recognition that could identify voice signal characteristics of a phone conversation, in which the facts could be classified as probably true or false. The results of the experiments done with the MFCC algorithm along with AI demonstrated that the extraction of the voice signal characteristics provided by the MFCC is consistent and permitted you to model a database to study the recognition of voice signal patterns. Thereby, it was possible to identify the user and the attendant"™s voice in a phone call, as well as identify the characteristics of the female, male and child"™s voice.

Author Biographies

Amivaldo Batista dos Santos, Instituto Federal Goiano

Mestre em Engenharia de Produção e Sistemas - PUC - Goiás. Graduado em Tecnologia em Processamento de Dados pelo Centro Universitário de Goiás - Uni-Anhanguera. Especialista em em Orientação a Objetos e Internet pela Uni Anhanguera. Atualmente professor no Instituto Federal Goiano.

Solange da Silva, Pontifícia Universidade Católica de Goiás

Doutora em Engenharia Elétrica pela Universidade Federal de Uberlândia. Mestre em Engenharia Elétrica e de Computação pela Universidade Federal de Goiás. Graduada em Ciências - Habilitação em Matemática pela Pontifícia Universidade Católica de Goiás, Especialista em Computação pela PUC Goiás. Atualmente professora na PUC Goiás.

Pedro José Abrão, Instituto Federal de Educação, Ciência e Tecnologia de Goiás

Doutor em Engenharia Elétrica pela Universidade Federal de Itajubá. Mestre Engenharia Elétrica pela Universidade Federal de Uberlândia. Graduado em Engenharia Elétrica pela Universidade Federal de Uberlândia. Atualmente professor titular do Instituto Federal de Educação, Ciência e Tecnologia de Goiás.

Luana Machado dos Santos, Centro Universitário Araguaia

Mestre em Engenharia de Produção e Sistemas pela PUC Goiás, Especialista em Gestão Empresarial com Ênfase em Consultoria pela UNIGOIÁS, Metodologias Ativas e Tecnologias Educacionais. Graduada em Administração pela PUC Goiás. Atualmente professora no Centro Universitário Araguaia.

Downloads

Published

2024-02-19

Issue

v. 18, n. 3 (2023)

Section

Articles

License

The copyright of the published articles will be transferred to the Uniaaraguaia Magazine, allowing its subsequent reproduction as transcription and with due citation of source. In the event of acceptance and before the publication of the article, the plaintiff (s) shall write a statement formally transferring copyright to the magazine.
The author may also print and distribute copies of his article, provided that he mentions that the rights belong to the Uniaaraguaia Magazine.

Author rights include the right to reproduce in full or partly by any means, distribute this article, including figures and photographs.

By submitting originals to the Uniaaraguaia magazine, the author or authors express agreement with the following terms:

a) Authors maintain copyright and grant Uniaraguaia magazine the right of first publication, with the work simultaneously licensed under the Creative Commons Attribution license that allows the sharing of work with recognition of the authorship and initial publication in this magazine.

b) Authors are authorized to assume additional contracts separately, for non-expiration distribution of the work version published in this magazine (eg publish in institutional repository or as book chapter), with recognition of authorship and initial publication in this journal.

c) Authors are allowed and are encouraged to publish and distribute their work online (eg in institutional repositories or on their personal page) to any point before or during the editorial process, as this can generate productive changes as well as increase the impact and citation of published work.