An HMM system for recognizing articulation features for arabic phones

Hossam Hammady, Osama Badawy, Sherif Abdou, Mohsen Rashwan

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, we introduce a Hidden Markov Model (HMM) recognition system for the articulation features of Arabic phones. The low-level features are described by Mel-Frequency Cepstral Coefficients (MFCCs). The created HMMs directly model certain articulation features (fricative and plosive). Classification is done on these features regardless of the phone itself. The model has been created successfully and tested on reference speech data. The error rate is very low for many phones and acceptable for most of them. Accordingly, the system output can be used as a confidence measure applied to other existing speech recognizers. Finally, the recognizer is speaker-independent and context-independent.

Original languageEnglish
Title of host publication2008 International Conference on Computer Engineering and Systems, ICCES 2008
Pages125-130
Number of pages6
DOIs
Publication statusPublished - 1 Dec 2008
Externally publishedYes
Event2008 International Conference on Computer Engineering and Systems, ICCES 2008 - Cairo, Egypt
Duration: 25 Nov 200827 Nov 2008

Other

Other2008 International Conference on Computer Engineering and Systems, ICCES 2008
CountryEgypt
CityCairo
Period25/11/0827/11/08

Fingerprint

Hidden Markov models

Keywords

  • Feature extraction
  • Hidden markov models
  • Speech recognition

ASJC Scopus subject areas

  • Computer Science Applications
  • Hardware and Architecture
  • Software

Cite this

Hammady, H., Badawy, O., Abdou, S., & Rashwan, M. (2008). An HMM system for recognizing articulation features for arabic phones. In 2008 International Conference on Computer Engineering and Systems, ICCES 2008 (pp. 125-130). [4772980] https://doi.org/10.1109/ICCES.2008.4772980

An HMM system for recognizing articulation features for arabic phones. / Hammady, Hossam; Badawy, Osama; Abdou, Sherif; Rashwan, Mohsen.

2008 International Conference on Computer Engineering and Systems, ICCES 2008. 2008. p. 125-130 4772980.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Hammady, H, Badawy, O, Abdou, S & Rashwan, M 2008, An HMM system for recognizing articulation features for arabic phones. in 2008 International Conference on Computer Engineering and Systems, ICCES 2008., 4772980, pp. 125-130, 2008 International Conference on Computer Engineering and Systems, ICCES 2008, Cairo, Egypt, 25/11/08. https://doi.org/10.1109/ICCES.2008.4772980
Hammady H, Badawy O, Abdou S, Rashwan M. An HMM system for recognizing articulation features for arabic phones. In 2008 International Conference on Computer Engineering and Systems, ICCES 2008. 2008. p. 125-130. 4772980 https://doi.org/10.1109/ICCES.2008.4772980
Hammady, Hossam ; Badawy, Osama ; Abdou, Sherif ; Rashwan, Mohsen. / An HMM system for recognizing articulation features for arabic phones. 2008 International Conference on Computer Engineering and Systems, ICCES 2008. 2008. pp. 125-130
@inproceedings{b6a0c437b9cb4b40b7d3cbe026bc26b9,
title = "An HMM system for recognizing articulation features for arabic phones",
abstract = "In this paper, we introduce a Hidden Markov Model (HMM) recognition system for the articulation features of Arabic phones. The low-level features are described by Mel-Frequency Cepstral Coefficients (MFCCs). The created HMMs directly model certain articulation features (fricative and plosive). Classification is done on these features regardless of the phone itself. The model has been created successfully and tested on reference speech data. The error rate is very low for many phones and acceptable for most of them. Accordingly, the system output can be used as a confidence measure applied to other existing speech recognizers. Finally, the recognizer is speaker-independent and context-independent.",
keywords = "Feature extraction, Hidden markov models, Speech recognition",
author = "Hossam Hammady and Osama Badawy and Sherif Abdou and Mohsen Rashwan",
year = "2008",
month = "12",
day = "1",
doi = "10.1109/ICCES.2008.4772980",
language = "English",
isbn = "9781424421152",
pages = "125--130",
booktitle = "2008 International Conference on Computer Engineering and Systems, ICCES 2008",

}

TY - GEN

T1 - An HMM system for recognizing articulation features for arabic phones

AU - Hammady, Hossam

AU - Badawy, Osama

AU - Abdou, Sherif

AU - Rashwan, Mohsen

PY - 2008/12/1

Y1 - 2008/12/1

N2 - In this paper, we introduce a Hidden Markov Model (HMM) recognition system for the articulation features of Arabic phones. The low-level features are described by Mel-Frequency Cepstral Coefficients (MFCCs). The created HMMs directly model certain articulation features (fricative and plosive). Classification is done on these features regardless of the phone itself. The model has been created successfully and tested on reference speech data. The error rate is very low for many phones and acceptable for most of them. Accordingly, the system output can be used as a confidence measure applied to other existing speech recognizers. Finally, the recognizer is speaker-independent and context-independent.

AB - In this paper, we introduce a Hidden Markov Model (HMM) recognition system for the articulation features of Arabic phones. The low-level features are described by Mel-Frequency Cepstral Coefficients (MFCCs). The created HMMs directly model certain articulation features (fricative and plosive). Classification is done on these features regardless of the phone itself. The model has been created successfully and tested on reference speech data. The error rate is very low for many phones and acceptable for most of them. Accordingly, the system output can be used as a confidence measure applied to other existing speech recognizers. Finally, the recognizer is speaker-independent and context-independent.

KW - Feature extraction

KW - Hidden markov models

KW - Speech recognition

UR - http://www.scopus.com/inward/record.url?scp=67649531763&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=67649531763&partnerID=8YFLogxK

U2 - 10.1109/ICCES.2008.4772980

DO - 10.1109/ICCES.2008.4772980

M3 - Conference contribution

SN - 9781424421152

SP - 125

EP - 130

BT - 2008 International Conference on Computer Engineering and Systems, ICCES 2008

ER -