Statistical analysis of Arabic phonemes used in Arabic speech recognition

Khalid M.O. Nahar; Mustafa Elshafei; Wasfi G. Al-Khatib; Husni Al-Muhtaseb; Mansour M. Alghamdi

doi:10.1007/978-3-642-34475-6_64

Statistical analysis of Arabic phonemes used in Arabic speech recognition

Khalid M.O. Nahar^*, Mustafa Elshafei, Wasfi G. Al-Khatib, Husni Al-Muhtaseb, Mansour M. Alghamdi

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

4 Scopus citations

Abstract

This study is specifically concerned with the statistical analysis of the Arabic phonemes due to its significant role in continuous Arabic Speech Recognition System (ASR). When building Arabic speech recognizer , the number of frames that a phoneme occupy, the phoneme boundary and the number of Hidden Markov Model necessary to represent the phoneme are greatly helpful in enhancing the recognition accuracy. In this paper we statically analyze KACST-5 hours corpus, which was used in Arabic speech recognition for both training and recognition. The results showed different set of tables and figures that are helpful for Arabic speech researchers. The paper comes up with a clustering graph for Arabic phonemes based on the median and a trigram table for all phonemes which represent the frequency of a phoneme to appear in trigram. The study was consistent and agreed with Arabic speech scientist's observations.

Original language	English
Title of host publication	Neural Information Processing - 19th International Conference, ICONIP 2012, Proceedings
Pages	533-542
Number of pages	10
Edition	PART 1
DOIs	https://doi.org/10.1007/978-3-642-34475-6_64
State	Published - 2012

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Number	PART 1
Volume	7663 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Keywords

Acoustic Model
Arabic Speech Recognition
HMM
Kacst Arabic speech corpus
MFCC files
Median
Mode
Phoneme
Standard Deviation
Variance

ASJC Scopus subject areas

Theoretical Computer Science
General Computer Science

Access to Document

10.1007/978-3-642-34475-6_64

Cite this

Nahar, K. M. O., Elshafei, M., Al-Khatib, W. G., Al-Muhtaseb, H., & Alghamdi, M. M. (2012). Statistical analysis of Arabic phonemes used in Arabic speech recognition. In Neural Information Processing - 19th International Conference, ICONIP 2012, Proceedings (PART 1 ed., pp. 533-542). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7663 LNCS, No. PART 1). https://doi.org/10.1007/978-3-642-34475-6_64

Nahar, Khalid M.O. ; Elshafei, Mustafa ; Al-Khatib, Wasfi G. et al. / Statistical analysis of Arabic phonemes used in Arabic speech recognition. Neural Information Processing - 19th International Conference, ICONIP 2012, Proceedings. PART 1. ed. 2012. pp. 533-542 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 1).

@inproceedings{49dd23b5ad594cb380e6fbabcd9f99ec,

title = "Statistical analysis of Arabic phonemes used in Arabic speech recognition",

abstract = "This study is specifically concerned with the statistical analysis of the Arabic phonemes due to its significant role in continuous Arabic Speech Recognition System (ASR). When building Arabic speech recognizer , the number of frames that a phoneme occupy, the phoneme boundary and the number of Hidden Markov Model necessary to represent the phoneme are greatly helpful in enhancing the recognition accuracy. In this paper we statically analyze KACST-5 hours corpus, which was used in Arabic speech recognition for both training and recognition. The results showed different set of tables and figures that are helpful for Arabic speech researchers. The paper comes up with a clustering graph for Arabic phonemes based on the median and a trigram table for all phonemes which represent the frequency of a phoneme to appear in trigram. The study was consistent and agreed with Arabic speech scientist's observations.",

keywords = "Acoustic Model, Arabic Speech Recognition, HMM, Kacst Arabic speech corpus, MFCC files, Median, Mode, Phoneme, Standard Deviation, Variance",

author = "Nahar, {Khalid M.O.} and Mustafa Elshafei and Al-Khatib, {Wasfi G.} and Husni Al-Muhtaseb and Alghamdi, {Mansour M.}",

year = "2012",

doi = "10.1007/978-3-642-34475-6_64",

language = "English",

isbn = "9783642344749",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

number = "PART 1",

pages = "533--542",

booktitle = "Neural Information Processing - 19th International Conference, ICONIP 2012, Proceedings",

edition = "PART 1",

}

Nahar, KMO, Elshafei, M, Al-Khatib, WG , Al-Muhtaseb, H & Alghamdi, MM 2012, Statistical analysis of Arabic phonemes used in Arabic speech recognition. in Neural Information Processing - 19th International Conference, ICONIP 2012, Proceedings. PART 1 edn, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), no. PART 1, vol. 7663 LNCS, pp. 533-542. https://doi.org/10.1007/978-3-642-34475-6_64

Statistical analysis of Arabic phonemes used in Arabic speech recognition. / Nahar, Khalid M.O.; Elshafei, Mustafa; Al-Khatib, Wasfi G. et al.
Neural Information Processing - 19th International Conference, ICONIP 2012, Proceedings. PART 1. ed. 2012. p. 533-542 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7663 LNCS, No. PART 1).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Statistical analysis of Arabic phonemes used in Arabic speech recognition

AU - Nahar, Khalid M.O.

AU - Elshafei, Mustafa

AU - Al-Khatib, Wasfi G.

AU - Al-Muhtaseb, Husni

AU - Alghamdi, Mansour M.

PY - 2012

Y1 - 2012

N2 - This study is specifically concerned with the statistical analysis of the Arabic phonemes due to its significant role in continuous Arabic Speech Recognition System (ASR). When building Arabic speech recognizer , the number of frames that a phoneme occupy, the phoneme boundary and the number of Hidden Markov Model necessary to represent the phoneme are greatly helpful in enhancing the recognition accuracy. In this paper we statically analyze KACST-5 hours corpus, which was used in Arabic speech recognition for both training and recognition. The results showed different set of tables and figures that are helpful for Arabic speech researchers. The paper comes up with a clustering graph for Arabic phonemes based on the median and a trigram table for all phonemes which represent the frequency of a phoneme to appear in trigram. The study was consistent and agreed with Arabic speech scientist's observations.

AB - This study is specifically concerned with the statistical analysis of the Arabic phonemes due to its significant role in continuous Arabic Speech Recognition System (ASR). When building Arabic speech recognizer , the number of frames that a phoneme occupy, the phoneme boundary and the number of Hidden Markov Model necessary to represent the phoneme are greatly helpful in enhancing the recognition accuracy. In this paper we statically analyze KACST-5 hours corpus, which was used in Arabic speech recognition for both training and recognition. The results showed different set of tables and figures that are helpful for Arabic speech researchers. The paper comes up with a clustering graph for Arabic phonemes based on the median and a trigram table for all phonemes which represent the frequency of a phoneme to appear in trigram. The study was consistent and agreed with Arabic speech scientist's observations.

KW - Acoustic Model

KW - Arabic Speech Recognition

KW - HMM

KW - Kacst Arabic speech corpus

KW - MFCC files

KW - Median

KW - Mode

KW - Phoneme

KW - Standard Deviation

KW - Variance

UR - http://www.scopus.com/inward/record.url?scp=84869029966&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-34475-6_64

DO - 10.1007/978-3-642-34475-6_64

M3 - Conference contribution

AN - SCOPUS:84869029966

SN - 9783642344749

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 533

EP - 542

BT - Neural Information Processing - 19th International Conference, ICONIP 2012, Proceedings

ER -

Nahar KMO, Elshafei M, Al-Khatib WG , Al-Muhtaseb H, Alghamdi MM. Statistical analysis of Arabic phonemes used in Arabic speech recognition. In Neural Information Processing - 19th International Conference, ICONIP 2012, Proceedings. PART 1 ed. 2012. p. 533-542. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 1). doi: 10.1007/978-3-642-34475-6_64

Statistical analysis of Arabic phonemes used in Arabic speech recognition

Abstract

Publication series

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this