A Discriminative Condition-Aware Backend for Speaker Verification | ||
| dc.conicet.organizacion | Consejo Nacional de Investigaciones Científicas y Técnicas. Oficina de Coordinación Administrativa Ciudad Universitaria. Instituto de Investigación en Ciencias de la Computación. Universidad de Buenos Aires. Facultad de Ciencias Exactas y Naturales. Instituto de Investigación en Ciencias de la Computación | |
| dc.conicet.organizacion | Sri International | |
| dc.conicet.rol | Autor | |
| dc.conicet.rol | Autor | |
| dc.contributor.author | Ferrer, Luciana | |
| dc.contributor.author | McLaren, Mitchell | |
| dc.coverage | Internacional | |
| dc.date.evento | 2020-05-04 | |
| dc.date.eventoHasta | 2020-05-08 | |
| dc.date.issued | 2020 | es |
| dc.description.abstract | We present a scoring approach for speaker verification that mimics the standard PLDA-based backend process used in most current speaker verification systems. However, unlike the standard backends, all parameters of the model are jointly trained to optimize the binary cross-entropy for the speaker verification task. We further integrate the calibration stage inside the model, making the parameters of this stage depend on metadata vectors that represent the conditions of the signals. We show that the proposed backend has excellent outof-the-box calibration performance on most of our test sets, making it an ideal approach for cases in which the test conditions are not known and development data is not available for training a domainspecific calibration model. | es |
| dc.description.ciudadEvento | Barcelona | es |
| dc.description.fil | Fil: Ferrer, Luciana. Consejo Nacional de Investigaciones Científicas y Técnicas. Oficina de Coordinación Administrativa Ciudad Universitaria. Instituto de Investigación en Ciencias de la Computación. Universidad de Buenos Aires. Facultad de Ciencias Exactas y Naturales. Instituto de Investigación en Ciencias de la Computación; Argentina | |
| dc.description.fil | Fil: McLaren, Mitchell. Sri International; Estados Unidos | |
| dc.description.institucionOrganizadora | Institute of Electrical and Electronics Engineers | es |
| dc.description.nombreEvento | 2020 IEEE International Conference on Acoustics, Speech and Signal Processing | es |
| dc.description.paisEvento | España | es |
| dc.format | application/pdf | es |
| dc.identifier.citation | A Discriminative Condition-Aware Backend for Speaker Verification; 2020 IEEE International Conference on Acoustics, Speech and Signal Processing; Barcelona; España; 2020; 6604-6608 | es |
| dc.identifier.eissn | 1520-6149 | es |
| dc.identifier.institucional | CONICET | |
| dc.identifier.issn | 2379-190X | es |
| dc.identifier.repo | CONICET Digital | |
| dc.identifier.uri | http://hdl.handle.net/11336/271761 | |
| dc.journal.pagination | 6604-6608 | es |
| dc.journal.pais | Estados Unidos | es |
| dc.language.iso | eng | es |
| dc.publisher | Institute of Electrical and Electronics Engineers | es |
| dc.relation.alternativeid | info:eu-repo/semantics/altIdentifier/url/https://ieeexplore.ieee.org/abstract/document/9053485 | |
| dc.rights | info:eu-repo/semantics/restrictedAccess | * |
| dc.rights.uri | https://creativecommons.org/licenses/by-nc-sa/2.5/ar/ | * |
| dc.source.revista | Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing | es |
| dc.subject | SPEAKER VERIFICATION | es |
| dc.subject | PROBABILISTIC LINEAR DISCRIMINANT ANALYSIS | es |
| dc.subject | CALIBRATION | es |
| dc.subject | CONDITION ROBUSTNESS | es |
| dc.subject.classification | Ciencias de la Información y Bioinformática | es |
| dc.subject.classification | Ciencias de la Computación e Información | es |
| dc.subject.classification | CIENCIAS NATURALES Y EXACTAS | es |
| dc.title | A Discriminative Condition-Aware Backend for Speaker Verification | es |
| dc.type | info:eu-repo/semantics/publishedVersion | |
| dc.type | info:eu-repo/semantics/conferenceObject | |
| dc.type | info:ar-repo/semantics/documento de conferencia | |
| dc.type.publicacion | Journal | es |
| dc.type.subtype | Conferencia | es |
Archivos
Bloque original
Cargando...
- Nombre:
- CONICET_Digital_Nro.f06e5bbe-b48b-47bd-9660-c2cf240846bc_B.pdf
- Tamaño:
- 211.67 KB
- Formato:
- Adobe Portable Document Format