Knowledge-Based Wavelet Filters Prominently Detect Spoofed Speech

Registro completo de metadados
MetadadosDescriçãoIdioma
Autor(es): dc.contributorUniversidade Estadual Paulista (UNESP)-
Autor(es): dc.creatorDe Almeida, Alex M. G.-
Autor(es): dc.creatorGuido, Rodrigo Capobianco-
Data de aceite: dc.date.accessioned2025-08-21T16:44:46Z-
Data de disponibilização: dc.date.available2025-08-21T16:44:46Z-
Data de envio: dc.date.issued2022-05-01-
Data de envio: dc.date.issued2022-05-01-
Data de envio: dc.date.issued2020-12-31-
Fonte completa do material: dc.identifierhttp://dx.doi.org/10.1109/ISM52913.2021.00014-
Fonte completa do material: dc.identifierhttp://hdl.handle.net/11449/234169-
Fonte: dc.identifier.urihttp://educapes.capes.gov.br/handle/11449/234169-
Descrição: dc.descriptionAlthough voice biometrics has been at the forefront of speech technologies, spoofing attacks have been one of the main issues responsible for avoiding its practical usage in commercial applications. Consequently, this article presents our proposed approach for building Knowledge-based wavelet filters, particularly dedicated to distinguish between genuine and spoofed speech. The main contribution of our strategy, particularly dedicated to identify replay attacks, is the acquisition of knowledge through the so called knowledge coefficients. Our results, obtained upon performing a set of experiments, indicate that the designed wavelet filters successfully classified hundreds of speech signals from our dataset, stimulating further research in this direction.-
Descrição: dc.descriptionSão Paulo State University São Jose Do Rio Preto-
Descrição: dc.descriptionSão Paulo State University São Jose Do Rio Preto-
Formato: dc.format35-38-
Idioma: dc.languageen-
Relação: dc.relationProceedings - 23rd IEEE International Symposium on Multimedia, ISM 2021-
???dc.source???: dc.sourceScopus-
Palavras-chave: dc.subjectSpeaker Verification-
Palavras-chave: dc.subjectWavelet-
Título: dc.titleKnowledge-Based Wavelet Filters Prominently Detect Spoofed Speech-
Tipo de arquivo: dc.typeaula digital-
Aparece nas coleções:Repositório Institucional - Unesp

Não existem arquivos associados a este item.