Result Details

BUT System Description to SdSV Challenge 2020

BURGET, L.; GLEMBEK, O.; LOZANO DÍEZ, A.; MATĚJKA, P.; NOVOTNÝ, O.; PLCHOT, O.; PULUGUNDLA, B.; ROHDIN, J.; SILNOVA, A.; VESELÝ, K. BUT System Description to SdSV Challenge 2020. Proceedings of Short-duration Speaker Verification Challenge 2020 Workshop. Shanghai, on-line event of Interspeech 2020 Conference: 2020. p. 1-5.
Type
conference paper
Language
English
Authors
Burget Lukáš, doc. Ing., Ph.D., DCGM (FIT)
Glembek Ondřej, Ing., Ph.D., DCGM (FIT)
Lozano Díez Alicia, Ph.D., DCGM (FIT)
Matějka Pavel, Ing., Ph.D., DCGM (FIT)
Novotný Ondřej, Ing., Ph.D., DCGM (FIT)
Plchot Oldřich, Ing., Ph.D., DCGM (FIT)
Pulugundla Bhargav, M.Sc., DCGM (FIT)
Rohdin Johan Andréas, M.Sc., Ph.D., FIT (FIT), DCGM (FIT)
Silnova Anna, M.Sc., Ph.D., DCGM (FIT)
Veselý Karel, Ing., Ph.D., DCGM (FIT)
Abstract

In this report, we describe the submission of Brno Universityof Technology (BUT) team to the Short Duration Speaker Verification(SdSV) Challenge 2020. For the text-dependent task,our primary submission consists of a simple linear logistic regressionscore level fusion of different i-vector and x-vectorbased systems. Our i-vector systems are based on concatenatedMFCC and bottleneck features. For both types of embeddings,we use PLDA backends, showing the success of phrasedependenttraining of PLDA and its combination with a Gaussianlinear classifier phrase recognizer. For the task of textindependentspeaker verification, we combine three different xvectorsystems based on TDNN and ResNet architectures.

Keywords

short duration speaker verification, phrasedependentPLDA, phrase recognizer, x-vector, TDNN, ResNet

URL
Published
2020
Pages
1–5
Proceedings
Proceedings of Short-duration Speaker Verification Challenge 2020 Workshop
Conference
Short-duration Speaker Verification (SdSV) Challenge 2020 Workshop, Shangai, CN
Place
Shanghai, on-line event of Interspeech 2020 Conference
BibTeX 
@inproceedings{BUT171000,
  author="Lukáš {Burget} and Ondřej {Glembek} and Alicia {Lozano Díez} and Pavel {Matějka} and Ondřej {Novotný} and Oldřich {Plchot} and Bhargav {Pulugundla} and Johan Andréas {Rohdin} and Anna {Silnova} and Karel {Veselý}",
  title="BUT System Description to SdSV Challenge 2020",
  booktitle="Proceedings of Short-duration Speaker Verification Challenge 2020 Workshop",
  year="2020",
  pages="1--5",
  address="Shanghai, on-line event of Interspeech 2020 Conference",
  url="https://sdsvc.github.io/2020/descriptions/Team56_Both.pdf"
}
Files
Projects
Improving Robustnes in Automatic Speaker Recognition, GACR, Juniorské granty, GJ17-23870Y, start: 2017-01-01, end: 2019-12-31, completed
Information mining in speech acquired by distant microphones, MV, Bezpečnostní výzkum České republiky 2015-2020, VI20152020025, start: 2015-10-01, end: 2020-09-30, completed
IT4Innovations excellence in science, MŠMT, Národní program udržitelnosti II, LQ1602, start: 2016-01-01, end: 2020-12-31, completed
Neural Representations in multi-modal and multi-lingual modeling, GACR, Grantové projekty exelence v základním výzkumu EXPRO - 2019, GX19-26934X, start: 2019-01-01, end: 2023-12-31, completed
Robust End-To-End SPEAKER recognition based on deep learning and attention models, EU, Horizon 2020, start: 2019-06-01, end: 2021-01-31, completed
Research groups
Speech Data Mining Research Group BUT Speech@FIT (RG SPEECH)
Departments
Department of Computer Graphics and Multimedia (DCGM)
Back to top