A bit more on the ability of adaptation of speech signals

Some traditional digital signal processing techniques encompass enhancement, filtering, coding, compression, detection and recognition. Recently, it has been presented a new hypothesis of signal processing known as the ability of adaptation of speech signals: an original speech signal may sound simi...

Descripción completa

Detalles Bibliográficos
Autores: Ballesteros, Dora Maria, Moreno Aróstegui, Juan Manuel|||0000-0001-5984-3596
Tipo de recurso: artículo
Fecha de publicación:2013
País:España
Institución:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/19502
Acceso en línea:https://hdl.handle.net/2117/19502
Access Level:acceso abierto
Palabra clave:Signal processing
Speech processing systems
Ability of adaptation
speech signals
wavelet coefficients
similarity
Tractament del senyal
Processament de la parla
Àrees temàtiques de la UPC::Enginyeria de la telecomunicació::Processament del senyal::Processament del senyal en les telecomunicacions
Àrees temàtiques de la UPC::Enginyeria de la telecomunicació::Processament del senyal::Filtres analògics i digitals
Descripción
Sumario:Some traditional digital signal processing techniques encompass enhancement, filtering, coding, compression, detection and recognition. Recently, it has been presented a new hypothesis of signal processing known as the ability of adaptation of speech signals: an original speech signal may sound similar to a target speech signal if a relocation process of its wavelet coefficients is applied. This hypothesis is true under some conditions theoretically defined. In this paper we present the basic idea behind the hypothesis of adaptation and moreover, we test the hypothesis within four cases: speech signals with the same gender and language, speech signals with the same gender but different language, speech signals with the same language but different gender, and speech signals with different gender and language. It is found that the hypothesis is true if the requirements are satis® ed, even if the gender or the language of the original and target signals are not the same.