Characterization of aspirated and unaspirated sounds in speech

Pravin Bhaskar Ramteke, Anmol Sadanand, Shashidhar G. Koolagudi, Vidya Pai

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

Abstract

In this work, consonant aspiration and unaspiration phenomena are studied. It is known that, pronunciation of aspiration and unaspiration is characterized by the 'puff of air' released at the place of constriction in the vocal tract which is known as burst. Here, the properties of vowel immediately after the burst are studied for characterization of the burst. Excitation source signal estimated from the speech linear prediction residual is used for the task. The signal characteristics such as glottal pulse, duration of open, closed & return phases, slope of open & return phases, duration of burst, ratio of highest and lowest energies of signal and voice onset time (VOT) are explored to characterize aspiration and unaspiration. TIMIT English speech corpus is used to test the proposed approach. Random forest (RF) and support vector machine (SVMs) are used as classifiers to test the effectiveness of the features used for the task. An accuracy of 99.93% and 94.03% is achieved respectively. From the results, it is observed that the proposed features are robust in classifying the aspirated and unaspirated consonants.

Original languageEnglish
Title of host publicationTENCON 2017 - 2017 IEEE Region 10 Conference
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages2840-2845
Number of pages6
Volume2017-December
ISBN (Electronic)9781509011339
DOIs
Publication statusPublished - 19-12-2017
Externally publishedYes
Event2017 IEEE Region 10 Conference, TENCON 2017 - Penang, Malaysia
Duration: 05-11-201708-11-2017

Conference

Conference2017 IEEE Region 10 Conference, TENCON 2017
CountryMalaysia
CityPenang
Period05-11-1708-11-17

Fingerprint

Acoustic waves
Support vector machines
Classifiers
Air

All Science Journal Classification (ASJC) codes

  • Computer Science Applications
  • Electrical and Electronic Engineering

Cite this

Ramteke, P. B., Sadanand, A., Koolagudi, S. G., & Pai, V. (2017). Characterization of aspirated and unaspirated sounds in speech. In TENCON 2017 - 2017 IEEE Region 10 Conference (Vol. 2017-December, pp. 2840-2845). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/TENCON.2017.8228345
Ramteke, Pravin Bhaskar ; Sadanand, Anmol ; Koolagudi, Shashidhar G. ; Pai, Vidya. / Characterization of aspirated and unaspirated sounds in speech. TENCON 2017 - 2017 IEEE Region 10 Conference. Vol. 2017-December Institute of Electrical and Electronics Engineers Inc., 2017. pp. 2840-2845
@inproceedings{f96c684be51d4d56b7a1db8fc8bf152b,
title = "Characterization of aspirated and unaspirated sounds in speech",
abstract = "In this work, consonant aspiration and unaspiration phenomena are studied. It is known that, pronunciation of aspiration and unaspiration is characterized by the 'puff of air' released at the place of constriction in the vocal tract which is known as burst. Here, the properties of vowel immediately after the burst are studied for characterization of the burst. Excitation source signal estimated from the speech linear prediction residual is used for the task. The signal characteristics such as glottal pulse, duration of open, closed & return phases, slope of open & return phases, duration of burst, ratio of highest and lowest energies of signal and voice onset time (VOT) are explored to characterize aspiration and unaspiration. TIMIT English speech corpus is used to test the proposed approach. Random forest (RF) and support vector machine (SVMs) are used as classifiers to test the effectiveness of the features used for the task. An accuracy of 99.93{\%} and 94.03{\%} is achieved respectively. From the results, it is observed that the proposed features are robust in classifying the aspirated and unaspirated consonants.",
author = "Ramteke, {Pravin Bhaskar} and Anmol Sadanand and Koolagudi, {Shashidhar G.} and Vidya Pai",
year = "2017",
month = "12",
day = "19",
doi = "10.1109/TENCON.2017.8228345",
language = "English",
volume = "2017-December",
pages = "2840--2845",
booktitle = "TENCON 2017 - 2017 IEEE Region 10 Conference",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
address = "United States",

}

Ramteke, PB, Sadanand, A, Koolagudi, SG & Pai, V 2017, Characterization of aspirated and unaspirated sounds in speech. in TENCON 2017 - 2017 IEEE Region 10 Conference. vol. 2017-December, Institute of Electrical and Electronics Engineers Inc., pp. 2840-2845, 2017 IEEE Region 10 Conference, TENCON 2017, Penang, Malaysia, 05-11-17. https://doi.org/10.1109/TENCON.2017.8228345

Characterization of aspirated and unaspirated sounds in speech. / Ramteke, Pravin Bhaskar; Sadanand, Anmol; Koolagudi, Shashidhar G.; Pai, Vidya.

TENCON 2017 - 2017 IEEE Region 10 Conference. Vol. 2017-December Institute of Electrical and Electronics Engineers Inc., 2017. p. 2840-2845.

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

TY - GEN

T1 - Characterization of aspirated and unaspirated sounds in speech

AU - Ramteke, Pravin Bhaskar

AU - Sadanand, Anmol

AU - Koolagudi, Shashidhar G.

AU - Pai, Vidya

PY - 2017/12/19

Y1 - 2017/12/19

N2 - In this work, consonant aspiration and unaspiration phenomena are studied. It is known that, pronunciation of aspiration and unaspiration is characterized by the 'puff of air' released at the place of constriction in the vocal tract which is known as burst. Here, the properties of vowel immediately after the burst are studied for characterization of the burst. Excitation source signal estimated from the speech linear prediction residual is used for the task. The signal characteristics such as glottal pulse, duration of open, closed & return phases, slope of open & return phases, duration of burst, ratio of highest and lowest energies of signal and voice onset time (VOT) are explored to characterize aspiration and unaspiration. TIMIT English speech corpus is used to test the proposed approach. Random forest (RF) and support vector machine (SVMs) are used as classifiers to test the effectiveness of the features used for the task. An accuracy of 99.93% and 94.03% is achieved respectively. From the results, it is observed that the proposed features are robust in classifying the aspirated and unaspirated consonants.

AB - In this work, consonant aspiration and unaspiration phenomena are studied. It is known that, pronunciation of aspiration and unaspiration is characterized by the 'puff of air' released at the place of constriction in the vocal tract which is known as burst. Here, the properties of vowel immediately after the burst are studied for characterization of the burst. Excitation source signal estimated from the speech linear prediction residual is used for the task. The signal characteristics such as glottal pulse, duration of open, closed & return phases, slope of open & return phases, duration of burst, ratio of highest and lowest energies of signal and voice onset time (VOT) are explored to characterize aspiration and unaspiration. TIMIT English speech corpus is used to test the proposed approach. Random forest (RF) and support vector machine (SVMs) are used as classifiers to test the effectiveness of the features used for the task. An accuracy of 99.93% and 94.03% is achieved respectively. From the results, it is observed that the proposed features are robust in classifying the aspirated and unaspirated consonants.

UR - http://www.scopus.com/inward/record.url?scp=85044194881&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85044194881&partnerID=8YFLogxK

U2 - 10.1109/TENCON.2017.8228345

DO - 10.1109/TENCON.2017.8228345

M3 - Conference contribution

VL - 2017-December

SP - 2840

EP - 2845

BT - TENCON 2017 - 2017 IEEE Region 10 Conference

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Ramteke PB, Sadanand A, Koolagudi SG, Pai V. Characterization of aspirated and unaspirated sounds in speech. In TENCON 2017 - 2017 IEEE Region 10 Conference. Vol. 2017-December. Institute of Electrical and Electronics Engineers Inc. 2017. p. 2840-2845 https://doi.org/10.1109/TENCON.2017.8228345