Context Independent Speaker Classiﬁcation

dc.contributor.author	Olshanetskyi, Borys
dc.date.accessioned	2020-02-25T15:33:50Z
dc.date.available	2020-02-25T15:33:50Z
dc.date.issued	2020
dc.identifier.citation	Olshanetskyi, Borys. Context Independent Speaker Classiﬁcation : Master Thesis : manuscript rights / Borys Olshanetskyi ; Supervisor Oleksii Molchanovskyi ; Ukrainian Catholic University, Department of Computer Sciences. – Lviv : [s.n.], 2020. – 36 p. : ill.	uk
dc.identifier.uri	http://er.ucu.edu.ua/handle/1/2052
dc.language.iso	en	uk
dc.subject	Mel Spectrogram	uk
dc.subject	Convolutional Neural Networks	uk
dc.subject	Convolutional Kernels	uk
dc.title	Context Independent Speaker Classiﬁcation	uk
dc.type	Preprint	uk
dc.status	Публікується вперше	uk
dc.description.abstracten	Speaker classiﬁcation is an essential task in the machine learning domain, with many practical applications in identiﬁcation and natural language processing. This work concentrates on speaker classiﬁcation as a subtask of general speaker diarization for real-world conversation scenarios. We research the domain of modern speech processing and present the original speaker classiﬁcation approach based on the recent developments in convolutional neural networks. Our method uses a spectrogram as input to the CNN classiﬁer model, allowing it to capture spatial information about voice frequencies distribution. Presented results show beyond human ability performance and give strong prospects for future development.	uk