Automatic discriminative measurement of voice onset time

Morgan Sonderegger; Joseph Keshet

doi:10.21437/interspeech.2010-616

Automatic discriminative measurement of voice onset time

Morgan Sonderegger, Joseph Keshet

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

10 Scopus citations

Abstract

We describe a discriminative algorithm for automatic VOT measurement, considered as an application of predicting structured output from speech. In contrast to previous studies which use customized rules, in our approach a function is trained on manually labeled examples, using an online algorithm to predict the burst and voicing onsets (and hence VOT). The feature set used is customized for detecting the burst and voicing onsets, and the loss function used in training is the difference between predicted and actual VOT. Applied to initial voiceless stops from two corpora, the algorithm compares favorably to previous work, and the agreement between automatic and manual measurements is near human inter-judge reliability.

Original language	English
Title of host publication	Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010
Publisher	International Speech Communication Association
Pages	2242-2245
Number of pages	4
DOIs	https://doi.org/10.21437/interspeech.2010-616
State	Published - 2010
Externally published	Yes

Publication series

Name	Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010

Keywords

Discriminative prediction
SVM
Structured prediction
Voice onset time

Access to Document

10.21437/interspeech.2010-616

Cite this

Sonderegger, M., & Keshet, J. (2010). Automatic discriminative measurement of voice onset time. In Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010 (pp. 2242-2245). (Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010). International Speech Communication Association. https://doi.org/10.21437/interspeech.2010-616

Sonderegger, Morgan ; Keshet, Joseph. / Automatic discriminative measurement of voice onset time. Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010. International Speech Communication Association, 2010. pp. 2242-2245 (Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010).

@inproceedings{27b9cee38f9b45eaa22bd205e717d6d9,

title = "Automatic discriminative measurement of voice onset time",

abstract = "We describe a discriminative algorithm for automatic VOT measurement, considered as an application of predicting structured output from speech. In contrast to previous studies which use customized rules, in our approach a function is trained on manually labeled examples, using an online algorithm to predict the burst and voicing onsets (and hence VOT). The feature set used is customized for detecting the burst and voicing onsets, and the loss function used in training is the difference between predicted and actual VOT. Applied to initial voiceless stops from two corpora, the algorithm compares favorably to previous work, and the agreement between automatic and manual measurements is near human inter-judge reliability.",

keywords = "Discriminative prediction, SVM, Structured prediction, Voice onset time",

author = "Morgan Sonderegger and Joseph Keshet",

year = "2010",

doi = "10.21437/interspeech.2010-616",

language = "אנגלית",

series = "Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010",

publisher = "International Speech Communication Association",

pages = "2242--2245",

booktitle = "Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010",

}

Sonderegger, M & Keshet, J 2010, Automatic discriminative measurement of voice onset time. in Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010. Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010, International Speech Communication Association, pp. 2242-2245. https://doi.org/10.21437/interspeech.2010-616

Automatic discriminative measurement of voice onset time. / Sonderegger, Morgan; Keshet, Joseph.
Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010. International Speech Communication Association, 2010. p. 2242-2245 (Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010).

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

TY - GEN

T1 - Automatic discriminative measurement of voice onset time

AU - Sonderegger, Morgan

AU - Keshet, Joseph

PY - 2010

Y1 - 2010

N2 - We describe a discriminative algorithm for automatic VOT measurement, considered as an application of predicting structured output from speech. In contrast to previous studies which use customized rules, in our approach a function is trained on manually labeled examples, using an online algorithm to predict the burst and voicing onsets (and hence VOT). The feature set used is customized for detecting the burst and voicing onsets, and the loss function used in training is the difference between predicted and actual VOT. Applied to initial voiceless stops from two corpora, the algorithm compares favorably to previous work, and the agreement between automatic and manual measurements is near human inter-judge reliability.

AB - We describe a discriminative algorithm for automatic VOT measurement, considered as an application of predicting structured output from speech. In contrast to previous studies which use customized rules, in our approach a function is trained on manually labeled examples, using an online algorithm to predict the burst and voicing onsets (and hence VOT). The feature set used is customized for detecting the burst and voicing onsets, and the loss function used in training is the difference between predicted and actual VOT. Applied to initial voiceless stops from two corpora, the algorithm compares favorably to previous work, and the agreement between automatic and manual measurements is near human inter-judge reliability.

KW - Discriminative prediction

KW - SVM

KW - Structured prediction

KW - Voice onset time

UR - https://www.scopus.com/pages/publications/79959831442

U2 - 10.21437/interspeech.2010-616

DO - 10.21437/interspeech.2010-616

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:79959831442

T3 - Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010

SP - 2242

EP - 2245

BT - Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010

PB - International Speech Communication Association

ER -

Sonderegger M, Keshet J. Automatic discriminative measurement of voice onset time. In Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010. International Speech Communication Association. 2010. p. 2242-2245. (Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010). doi: 10.21437/interspeech.2010-616

Automatic discriminative measurement of voice onset time

Abstract

Publication series

Keywords

Access to Document

Other files and links

Fingerprint

Cite this