Detection of the glottal closure by jumps in the statistical properties of the signal

R. Di Francesco; E. Moulines

Detection of the glottal closure by jumps in the statistical properties of the signal

R. Di Francesco
, E. Moulines

Orange Labs

Research output: Contribution to conference › Paper › peer-review

Abstract

Two new methods for the detection of the instant of glottal closur·e directly from the speech waveform are presented here. ~~th detect the abrupt changes in the slwrt-term spectral characterlstzcs of the speech signal wirhin a pitch period caused by glottal events. The same statistical approach of sequential detection of events by hypothesis testing is used for this purpose. The first method is based on the maximization of a likelihood ratio. the second one uses a divergence convexity test. Experiments on real speech demonstrate the improved robustness ofthese methods.

Original language	English
Pages	2039-2042
Number of pages	4
Publication status	Published - 1 Jan 1989
Externally published	Yes
Event	1st European Conference on Speech Communication and Technology, EUROSPEECH 1989 - Paris, France Duration: 27 Sept 1989 → 29 Sept 1989

Conference

Conference	1st European Conference on Speech Communication and Technology, EUROSPEECH 1989
Country/Territory	France
City	Paris
Period	27/09/89 → 29/09/89

Cite this

@conference{5392752f1be0454580a1e1674de36ae1,

title = "Detection of the glottal closure by jumps in the statistical properties of the signal",

abstract = "Two new methods for the detection of the instant of glottal closur·e directly from the speech waveform are presented here. \textasciitilde{}\textasciitilde{}th detect the abrupt changes in the slwrt-term spectral characterlstzcs of the speech signal wirhin a pitch period caused by glottal events. The same statistical approach of sequential detection of events by hypothesis testing is used for this purpose. The first method is based on the maximization of a likelihood ratio. the second one uses a divergence convexity test. Experiments on real speech demonstrate the improved robustness ofthese methods.",

author = "\{Di Francesco\}, R. and E. Moulines",

note = "Publisher Copyright: {\textcopyright} 1st European Conference on Speech Communication and Technology, EUROSPEECH 1989.; 1st European Conference on Speech Communication and Technology, EUROSPEECH 1989 ; Conference date: 27-09-1989 Through 29-09-1989",

year = "1989",

month = jan,

day = "1",

language = "English",

pages = "2039--2042",

}

TY - CONF

T1 - Detection of the glottal closure by jumps in the statistical properties of the signal

AU - Di Francesco, R.

AU - Moulines, E.

PY - 1989/1/1

Y1 - 1989/1/1

N2 - Two new methods for the detection of the instant of glottal closur·e directly from the speech waveform are presented here. ~~th detect the abrupt changes in the slwrt-term spectral characterlstzcs of the speech signal wirhin a pitch period caused by glottal events. The same statistical approach of sequential detection of events by hypothesis testing is used for this purpose. The first method is based on the maximization of a likelihood ratio. the second one uses a divergence convexity test. Experiments on real speech demonstrate the improved robustness ofthese methods.

AB - Two new methods for the detection of the instant of glottal closur·e directly from the speech waveform are presented here. ~~th detect the abrupt changes in the slwrt-term spectral characterlstzcs of the speech signal wirhin a pitch period caused by glottal events. The same statistical approach of sequential detection of events by hypothesis testing is used for this purpose. The first method is based on the maximization of a likelihood ratio. the second one uses a divergence convexity test. Experiments on real speech demonstrate the improved robustness ofthese methods.

M3 - Paper

AN - SCOPUS:85019573495

SP - 2039

EP - 2042

T2 - 1st European Conference on Speech Communication and Technology, EUROSPEECH 1989

Y2 - 27 September 1989 through 29 September 1989

ER -

Detection of the glottal closure by jumps in the statistical properties of the signal

Abstract

Conference

Fingerprint

Cite this