BUT Text-Dependent Speaker Verification System for SdSV Challenge 2020

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F20%3APU138636" target="_blank" >RIV/00216305:26230/20:PU138636 - isvavai.cz</a>

Výsledek na webu

<a href="https://www.isca-speech.org/archive/Interspeech_2020/pdfs/2882.pdf" target="_blank" >https://www.isca-speech.org/archive/Interspeech_2020/pdfs/2882.pdf</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.21437/Interspeech.2020-2882" target="_blank" >10.21437/Interspeech.2020-2882</a>

Jazyk výsledku

angličtina

Název v původním jazyce

BUT Text-Dependent Speaker Verification System for SdSV Challenge 2020

Popis výsledku v původním jazyce

In this paper, we present the winning BUT submission for the text-dependent task of the SdSV challenge 2020. Given the large amount of training data available in this challenge, we explore successful techniques from text-independent systems in the text-dependent scenario. In particular, we trained x-vector extractors on both in-domain and out-of-domain datasets and combine them with i-vectors trained on concatenated MFCCs and bottleneck features, which have proven effective for the text-dependent scenario. Moreover, we proposed the use of phrase-dependent PLDA backend for scoring and its combination with a simple phrase recognizer, which brings up to 63% relative improvement on our development set with respect to using standard PLDA. Finally, we combine our different i-vector and x-vector based systems using a simple linear logistic regression score level fusion, which provides 28% relative improvement on the evaluation set with respect to our best single system.

Název v anglickém jazyce

BUT Text-Dependent Speaker Verification System for SdSV Challenge 2020

Popis výsledku anglicky

In this paper, we present the winning BUT submission for the text-dependent task of the SdSV challenge 2020. Given the large amount of training data available in this challenge, we explore successful techniques from text-independent systems in the text-dependent scenario. In particular, we trained x-vector extractors on both in-domain and out-of-domain datasets and combine them with i-vectors trained on concatenated MFCCs and bottleneck features, which have proven effective for the text-dependent scenario. Moreover, we proposed the use of phrase-dependent PLDA backend for scoring and its combination with a simple phrase recognizer, which brings up to 63% relative improvement on our development set with respect to using standard PLDA. Finally, we combine our different i-vector and x-vector based systems using a simple linear logistic regression score level fusion, which provides 28% relative improvement on the evaluation set with respect to our best single system.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

Výsledek vznikl pri realizaci vícero projektů. Více informací v záložce Projekty.

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2020

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH

ISBN

—

ISSN

1990-9772

e-ISSN

—

Počet stran výsledku

5

Strana od-do

761-765

Název nakladatele

International Speech Communication Association

Místo vydání

Shanghai

Místo konání akce

Sanghai

Datum konání akce

25. 10. 2020

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

000833594100158