Phone boundary detection using sample-based acoustic parameters

You Yu Lin; Yih-Ru Wang; Yuan Fu Liao

Phone boundary detection using sample-based acoustic parameters

You Yu Lin^*, Yih-Ru Wang, Yuan Fu Liao

^*Corresponding author for this work

Department of Electrical and Computer Engineering

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

3 Scopus citations

Abstract

A sample-based phone boundary detection algorithm is proposed in this paper. Some sample-based acoustic parameters are first extracted in the proposed method, including six sub-band signal envelopes, sample-based KL distance and spectral entropy. Then, the sample-based KL distance is used for boundary candidates preselection. Last, a supervised neural network is employed for final boundary detection. Experimental results using the TIMIT speech corpus showed that EERs of 13.2% and 15.1% were achieved for the training and test data sets, respectively. Moreover, 43.5% and 88.2% of boundaries detected were within 80- and 240-sample error tolerance from manual labeling results at the EER operating point.

Original language	English
Pages	1397-1400
Number of pages	4
State	Published - Sep 2010
Event	11th Annual Conference of the International Speech Communication Association: Spoken Language Processing for All, INTERSPEECH 2010 - Makuhari, Chiba, Japan Duration: 26 Sep 2010 → 30 Sep 2010

Conference

Conference	11th Annual Conference of the International Speech Communication Association: Spoken Language Processing for All, INTERSPEECH 2010
Country/Territory	Japan
City	Makuhari, Chiba
Period	26/09/10 → 30/09/10

Keywords

Speech analysis
Speech segmentation

Cite this

@conference{d9b27300e82e46ddaa36cec609129916,

title = "Phone boundary detection using sample-based acoustic parameters",

abstract = "A sample-based phone boundary detection algorithm is proposed in this paper. Some sample-based acoustic parameters are first extracted in the proposed method, including six sub-band signal envelopes, sample-based KL distance and spectral entropy. Then, the sample-based KL distance is used for boundary candidates preselection. Last, a supervised neural network is employed for final boundary detection. Experimental results using the TIMIT speech corpus showed that EERs of 13.2% and 15.1% were achieved for the training and test data sets, respectively. Moreover, 43.5% and 88.2% of boundaries detected were within 80- and 240-sample error tolerance from manual labeling results at the EER operating point.",

keywords = "Speech analysis, Speech segmentation",

author = "Lin, {You Yu} and Yih-Ru Wang and Liao, {Yuan Fu}",

year = "2010",

month = sep,

language = "English",

pages = "1397--1400",

note = "11th Annual Conference of the International Speech Communication Association: Spoken Language Processing for All, INTERSPEECH 2010 ; Conference date: 26-09-2010 Through 30-09-2010",

}

TY - CONF

T1 - Phone boundary detection using sample-based acoustic parameters

AU - Lin, You Yu

AU - Wang, Yih-Ru

AU - Liao, Yuan Fu

PY - 2010/9

Y1 - 2010/9

N2 - A sample-based phone boundary detection algorithm is proposed in this paper. Some sample-based acoustic parameters are first extracted in the proposed method, including six sub-band signal envelopes, sample-based KL distance and spectral entropy. Then, the sample-based KL distance is used for boundary candidates preselection. Last, a supervised neural network is employed for final boundary detection. Experimental results using the TIMIT speech corpus showed that EERs of 13.2% and 15.1% were achieved for the training and test data sets, respectively. Moreover, 43.5% and 88.2% of boundaries detected were within 80- and 240-sample error tolerance from manual labeling results at the EER operating point.

AB - A sample-based phone boundary detection algorithm is proposed in this paper. Some sample-based acoustic parameters are first extracted in the proposed method, including six sub-band signal envelopes, sample-based KL distance and spectral entropy. Then, the sample-based KL distance is used for boundary candidates preselection. Last, a supervised neural network is employed for final boundary detection. Experimental results using the TIMIT speech corpus showed that EERs of 13.2% and 15.1% were achieved for the training and test data sets, respectively. Moreover, 43.5% and 88.2% of boundaries detected were within 80- and 240-sample error tolerance from manual labeling results at the EER operating point.

KW - Speech analysis

KW - Speech segmentation

UR - http://www.scopus.com/inward/record.url?scp=79959825632&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:79959825632

SP - 1397

EP - 1400

T2 - 11th Annual Conference of the International Speech Communication Association: Spoken Language Processing for All, INTERSPEECH 2010

Y2 - 26 September 2010 through 30 September 2010

ER -

Phone boundary detection using sample-based acoustic parameters

Abstract

Conference

Keywords

Other files and links

Fingerprint

Cite this