Biometria de voz com base na replicação do pitch

• Ano de defesa: 2018
• Autor(a) principal: Moreno, Luis Carlos
• Orientador(a): Lopes, Paulo Batista
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Presbiteriana Mackenzie
• Palavras-chave em Português: autenticação; biometria; pitch; algoritmo; replicação do pitch
• Área do conhecimento CNPq: CNPQ::ENGENHARIAS::ENGENHARIA ELETRICA::TELECOMUNICACOES::SISTEMAS DE TELECOMUNICACOES
• Link de acesso: http://dspace.mackenzie.br/handle/10899/24490

Authentication and security in automated systems have become very much necessary in our days and many techniques have been proposed towards this end. One of these alternatives is biometrics in which human body characteristics are used to authenticate the system user. The objective of this work is to present a method of text independent speaker identi cation through the replication of pitch characteristics. Pitch is an important speech feature and is used in a variety of applications, including voice biometrics. The proposed method of speaker identi cation is based on short records segments of speech, namely, three seconds for training and three seconds for the speaker determination. From these segments pitch characteristics are extracted and are used in the proposed method of replication for identi cation of the speaker, resulting in a reduction in the Equal error rate (EER) values.