Detail publikace

Analysis of BUT-PT Submission for NIST LRE 2017

PLCHOT, O. MATĚJKA, P. NOVOTNÝ, O. CUMANI, S. LOZANO DÍEZ, A. SLAVÍČEK, J. DIEZ SÁNCHEZ, M. GRÉZL, F. GLEMBEK, O. KAMSALI VEERA, M. SILNOVA, A. BURGET, L. ONDEL, L. KESIRAJU, S. ROHDIN, J.

Originální název

Analysis of BUT-PT Submission for NIST LRE 2017

Anglický název

Analysis of BUT-PT Submission for NIST LRE 2017

Jazyk

en

Originální abstrakt

In this paper, we summarize our efforts in the NIST Language Recognition Evaluations (LRE) 2017 which resulted in systems providing very competitive and state-of-the-art performance. We provide both the descriptions and the analysis of the systems that we included in our submission. We explain our partitioning of the datasets that we were provided by NIST for training and development, and we follow by describing the features, DNN models and classifiers that were used to produce the final systems. After covering the architecture of our submission, we concentrate on post-evaluation analysis. We compare different DNN Bottle-Neck features, i-vector systems of different sizes and architectures, different classifiers and we present experimental results with data augmentation and with improved architecture of the system based on DNN embeddings. We present the performance of the systems in the Fixed condition (where participants are required to use only predefined data sets) and in addition to official NIST LRE17 evaluation set, we also provide results on our internal development set which can serve as a baseline for other researchers, since all training data are fixed and provided by NIST.

Anglický abstrakt

In this paper, we summarize our efforts in the NIST Language Recognition Evaluations (LRE) 2017 which resulted in systems providing very competitive and state-of-the-art performance. We provide both the descriptions and the analysis of the systems that we included in our submission. We explain our partitioning of the datasets that we were provided by NIST for training and development, and we follow by describing the features, DNN models and classifiers that were used to produce the final systems. After covering the architecture of our submission, we concentrate on post-evaluation analysis. We compare different DNN Bottle-Neck features, i-vector systems of different sizes and architectures, different classifiers and we present experimental results with data augmentation and with improved architecture of the system based on DNN embeddings. We present the performance of the systems in the Fixed condition (where participants are required to use only predefined data sets) and in addition to official NIST LRE17 evaluation set, we also provide results on our internal development set which can serve as a baseline for other researchers, since all training data are fixed and provided by NIST.

Dokumenty

BibTex


@inproceedings{BUT155068,
  author="Oldřich {Plchot} and Pavel {Matějka} and Ondřej {Novotný} and Sandro {Cumani} and Alicia {Lozano Díez} and Mireia {Diez Sánchez} and František {Grézl} and Ondřej {Glembek} and Mounika {Kamsali Veera} and Anna {Silnova} and Lukáš {Burget} and Lucas Antoine Francois {Ondel} and Santosh {Kesiraju} and Johan Andréas {Rohdin}",
  title="Analysis of BUT-PT Submission for NIST LRE 2017",
  annote="In this paper, we summarize our efforts in the NIST Language Recognition
Evaluations (LRE) 2017 which resulted in systems providing very competitive and
state-of-the-art performance. We provide both the descriptions and the analysis
of the systems that we included in our submission. We explain our partitioning of
the datasets that we were provided by NIST for training and development, and we
follow by describing the features, DNN models and classifiers that were used to
produce the final systems. After covering the architecture of our submission, we
concentrate on post-evaluation analysis. We compare different DNN Bottle-Neck
features, i-vector systems of different sizes and architectures, different
classifiers and we present experimental results with data augmentation and with
improved architecture of the system based on DNN embeddings. We present the
performance of the systems in the Fixed condition (where participants are
required to use only predefined data sets) and in addition to official NIST LRE17
evaluation set, we also provide results on our internal development set which can
serve as a baseline for other researchers, since all training data are fixed and
provided by NIST.",
  address="International Speech Communication Association",
  booktitle="Proceedings of Odyssey 2018 The Speaker and Language Recognition Workshop",
  chapter="155068",
  doi="10.21437/Odyssey.2018-7",
  edition="NEUVEDEN",
  howpublished="online",
  institution="International Speech Communication Association",
  number="6",
  year="2018",
  month="june",
  pages="47--53",
  publisher="International Speech Communication Association",
  type="conference paper"
}