Porovnání několika procedur verifikace řečníka založených na GMM

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F49777513%3A23520%2F03%3A00000337" target="_blank" >RIV/49777513:23520/03:00000337 - isvavai.cz</a>

Nalezeny alternativní kódy

RIV/49777513:23520/03:00000074

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

On the Amount of Speech Data Necessary for Successful Speaker Identification

Popis výsledku v původním jazyce

The paper deals with the dependence between the speaker identification performance and the amount of test data. Three speaker identification procedures based on hidden Markov models (HMMs) of phonemes are presented here. One, which is quite commonly usedin the speaker recognition systems based on HMMs, uses the likelihood of the whole utterance for speaker identification. The other two that are proposed in this paper are based on the majority voting rule. The experiments were performed for two different situations: either both training and test data were obtained from the same channel, or they were obtained from different channels. All experiments show that the proposed speaker identification procedure based on the majority voting rule for sequences of phonemes allows us to reduce the amount of test data necessary for successful speaker identification.

Název v anglickém jazyce

On the Amount of Speech Data Necessary for Successful Speaker Identification

Popis výsledku anglicky

The paper deals with the dependence between the speaker identification performance and the amount of test data. Three speaker identification procedures based on hidden Markov models (HMMs) of phonemes are presented here. One, which is quite commonly usedin the speaker recognition systems based on HMMs, uses the likelihood of the whole utterance for speaker identification. The other two that are proposed in this paper are based on the majority voting rule. The experiments were performed for two different situations: either both training and test data were obtained from the same channel, or they were obtained from different channels. All experiments show that the proposed speaker identification procedure based on the majority voting rule for sequences of phonemes allows us to reduce the amount of test data necessary for successful speaker identification.

Druh

J_x - Nezařazeno - Článek v odborném periodiku (Jimp, Jsc a Jost)

CEP obor

JD - Využití počítačů, robotika a její aplikace

OECD FORD obor

—

Projekt

<a href="/cs/project/GA102%2F02%2F0124" target="_blank" >GA102/02/0124: Hlasové technologie v podpoře informační společnosti</a><br>

Návaznosti

Z - Vyzkumny zamer (s odkazem do CEZ)

Rok uplatnění

2003

Kód důvěrnosti údajů

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

Název periodika

Eurospeech

ISSN

1018-4074

e-ISSN

—

Svazek periodika

2003

Číslo periodika v rámci svazku

—

Stát vydavatele periodika

CH - Švýcarská konfederace

Počet stran výsledku

4

Strana od-do

3021

Kód UT WoS článku

—

EID výsledku v databázi Scopus

—