Gaussian meta-embeddings for efficient scoring of a heavy-tailed PLDA model

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F18%3APU130769" target="_blank" >RIV/00216305:26230/18:PU130769 - isvavai.cz</a>

Výsledek na webu

<a href="https://www.fit.vut.cz/research/publication/11790/" target="_blank" >https://www.fit.vut.cz/research/publication/11790/</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.21437/Odyssey.2018-49" target="_blank" >10.21437/Odyssey.2018-49</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Gaussian meta-embeddings for efficient scoring of a heavy-tailed PLDA model

Popis výsledku v původním jazyce

Embeddings in machine learning are low-dimensional representations of complex input patterns, with the property that simple geometric operations like Euclidean distances and dot products can be used for classification and comparison tasks. We introduce meta-embeddings, which live in more general inner product spaces and which are designed to better propagate uncertainty through the embedding bottleneck. Traditional embeddings are trained to maximize between-class and minimize within-class distances. Meta-embeddings are trained to maximize relevant information throughput. As a proof of concept in speaker recognition, we derive an extractor from the familiar generative Gaussian PLDA model (GPLDA). We show that GPLDA likelihood ratio scores are given by Hilbert space inner products between Gaussian likelihood functions, which we term Gaussian meta-embeddings (GMEs). Meta-embedding extractors can be generatively or discriminatively trained. GMEs extracted by GPLDA have fixed precisions and do not propagate uncertainty. We show that a generalization to heavy-tailed PLDA gives GMEs with variable precisions, which do propagate uncertainty. Experiments on NIST SRE 2010 and 2016 show that the proposed method applied to i-vectors without length normalization is up to 20% more accurate than GPLDA applied to length-normalized i-vectors.

Název v anglickém jazyce

Gaussian meta-embeddings for efficient scoring of a heavy-tailed PLDA model

Popis výsledku anglicky

Embeddings in machine learning are low-dimensional representations of complex input patterns, with the property that simple geometric operations like Euclidean distances and dot products can be used for classification and comparison tasks. We introduce meta-embeddings, which live in more general inner product spaces and which are designed to better propagate uncertainty through the embedding bottleneck. Traditional embeddings are trained to maximize between-class and minimize within-class distances. Meta-embeddings are trained to maximize relevant information throughput. As a proof of concept in speaker recognition, we derive an extractor from the familiar generative Gaussian PLDA model (GPLDA). We show that GPLDA likelihood ratio scores are given by Hilbert space inner products between Gaussian likelihood functions, which we term Gaussian meta-embeddings (GMEs). Meta-embedding extractors can be generatively or discriminatively trained. GMEs extracted by GPLDA have fixed precisions and do not propagate uncertainty. We show that a generalization to heavy-tailed PLDA gives GMEs with variable precisions, which do propagate uncertainty. Experiments on NIST SRE 2010 and 2016 show that the proposed method applied to i-vectors without length normalization is up to 20% more accurate than GPLDA applied to length-normalized i-vectors.

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

2018

Kód důvěrnosti údajů

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

Název statě ve sborníku

Proceedings of Odyssey 2018

ISBN

—

ISSN

2312-2846

e-ISSN

—

Počet stran výsledku

8

Strana od-do

349-356

Název nakladatele

International Speech Communication Association

Místo vydání

Les Sables d'Olonne

Místo konání akce

Les Sables d'Olonne, France

Datum konání akce

26. 6. 2018

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—