Competing models-based text-prompted speaker independent verification algorithm

Heungkyu Lee, Hanseok Ko

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

In this paper, we propose competing models based on text-prompted speaker independent verification algorithm for an intelligent surveillance guard robot, wherein a robot prompts a code (i.e. word or phrase) for user entrance authentication. The proposed system requires the text-prompted speaker independent verification. In addition, it does not require a speaker dependent model and extra trained model as an alternative hypothesis for log-likelihood ratio test because of memory limitation. This is due to the given application scenario that an administrator changes the voice code every day for security reasoning and the targeting domain is unlimited. To resolve these issues, we propose to exploit the sub-word based anti-models for log-likelihood normalization through reusing an acoustic model and competing with voice code model. Anti-models using automatic production rules are set up in an initial time by using the statistical distance of phonemes against a voice code. The proposed system uses a two-pass strategy using a SCHMM-based recognition and verification step. In addition, a harmonics-based spectral subtraction algorithm is applied for a noisy robustness on an outdoor environment. The performance evaluation is done by using a common Korean database, PBW452DB, which consists of 63,280 utterances of 452 isolated words recorded in silent environment.

Original languageEnglish
Pages (from-to)28-44
Number of pages17
JournalSpeech Communication
Volume48
Issue number1
DOIs
Publication statusPublished - 2006 Jan 1

Fingerprint

Model-based
Administrative Personnel
Acoustics
Robot
robot
Subword
Production Rules
Acoustic Model
User Authentication
Log-likelihood Ratio
Model
Databases
Subtraction
Likelihood Ratio Test
Robots
Surveillance
Normalization
Performance Evaluation
Resolve
Likelihood

Keywords

  • LRT
  • Utterance verification
  • Voice code verification

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering
  • Experimental and Cognitive Psychology
  • Linguistics and Language

Cite this

Competing models-based text-prompted speaker independent verification algorithm. / Lee, Heungkyu; Ko, Hanseok.

In: Speech Communication, Vol. 48, No. 1, 01.01.2006, p. 28-44.

Research output: Contribution to journalArticle

@article{b414eaa3e746407fb29043e1d72fc319,
title = "Competing models-based text-prompted speaker independent verification algorithm",
abstract = "In this paper, we propose competing models based on text-prompted speaker independent verification algorithm for an intelligent surveillance guard robot, wherein a robot prompts a code (i.e. word or phrase) for user entrance authentication. The proposed system requires the text-prompted speaker independent verification. In addition, it does not require a speaker dependent model and extra trained model as an alternative hypothesis for log-likelihood ratio test because of memory limitation. This is due to the given application scenario that an administrator changes the voice code every day for security reasoning and the targeting domain is unlimited. To resolve these issues, we propose to exploit the sub-word based anti-models for log-likelihood normalization through reusing an acoustic model and competing with voice code model. Anti-models using automatic production rules are set up in an initial time by using the statistical distance of phonemes against a voice code. The proposed system uses a two-pass strategy using a SCHMM-based recognition and verification step. In addition, a harmonics-based spectral subtraction algorithm is applied for a noisy robustness on an outdoor environment. The performance evaluation is done by using a common Korean database, PBW452DB, which consists of 63,280 utterances of 452 isolated words recorded in silent environment.",
keywords = "LRT, Utterance verification, Voice code verification",
author = "Heungkyu Lee and Hanseok Ko",
year = "2006",
month = "1",
day = "1",
doi = "10.1016/j.specom.2005.05.014",
language = "English",
volume = "48",
pages = "28--44",
journal = "Speech Communication",
issn = "0167-6393",
publisher = "Elsevier",
number = "1",

}

TY - JOUR

T1 - Competing models-based text-prompted speaker independent verification algorithm

AU - Lee, Heungkyu

AU - Ko, Hanseok

PY - 2006/1/1

Y1 - 2006/1/1

N2 - In this paper, we propose competing models based on text-prompted speaker independent verification algorithm for an intelligent surveillance guard robot, wherein a robot prompts a code (i.e. word or phrase) for user entrance authentication. The proposed system requires the text-prompted speaker independent verification. In addition, it does not require a speaker dependent model and extra trained model as an alternative hypothesis for log-likelihood ratio test because of memory limitation. This is due to the given application scenario that an administrator changes the voice code every day for security reasoning and the targeting domain is unlimited. To resolve these issues, we propose to exploit the sub-word based anti-models for log-likelihood normalization through reusing an acoustic model and competing with voice code model. Anti-models using automatic production rules are set up in an initial time by using the statistical distance of phonemes against a voice code. The proposed system uses a two-pass strategy using a SCHMM-based recognition and verification step. In addition, a harmonics-based spectral subtraction algorithm is applied for a noisy robustness on an outdoor environment. The performance evaluation is done by using a common Korean database, PBW452DB, which consists of 63,280 utterances of 452 isolated words recorded in silent environment.

AB - In this paper, we propose competing models based on text-prompted speaker independent verification algorithm for an intelligent surveillance guard robot, wherein a robot prompts a code (i.e. word or phrase) for user entrance authentication. The proposed system requires the text-prompted speaker independent verification. In addition, it does not require a speaker dependent model and extra trained model as an alternative hypothesis for log-likelihood ratio test because of memory limitation. This is due to the given application scenario that an administrator changes the voice code every day for security reasoning and the targeting domain is unlimited. To resolve these issues, we propose to exploit the sub-word based anti-models for log-likelihood normalization through reusing an acoustic model and competing with voice code model. Anti-models using automatic production rules are set up in an initial time by using the statistical distance of phonemes against a voice code. The proposed system uses a two-pass strategy using a SCHMM-based recognition and verification step. In addition, a harmonics-based spectral subtraction algorithm is applied for a noisy robustness on an outdoor environment. The performance evaluation is done by using a common Korean database, PBW452DB, which consists of 63,280 utterances of 452 isolated words recorded in silent environment.

KW - LRT

KW - Utterance verification

KW - Voice code verification

UR - http://www.scopus.com/inward/record.url?scp=28644440058&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=28644440058&partnerID=8YFLogxK

U2 - 10.1016/j.specom.2005.05.014

DO - 10.1016/j.specom.2005.05.014

M3 - Article

VL - 48

SP - 28

EP - 44

JO - Speech Communication

JF - Speech Communication

SN - 0167-6393

IS - 1

ER -