A rule-based backchannel prediction model using pitch and pause information

Khiet Phuong Truong; Ronald Walter Poppe; Dirk K.J. Heylen

A rule-based backchannel prediction model using pitch and pause information

Khiet Phuong Truong, Ronald Walter Poppe, Dirk K.J. Heylen

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

47 Citations (Scopus)

185 Downloads (Pure)

Abstract

We manually designed rules for a backchannel (BC) prediction model based on pitch and pause information. In short, the model predicts a BC when there is a pause of a certain length that is preceded by a falling or rising pitch. This model was validated against the Dutch IFADV Corpus in a corpus-based evaluation method. The results showed that our model performs slightly better than another well-known rule-based BC prediction model that uses only pitch information. We observed that the length of a pause preceding a BC is one of the important features in this model, next to the duration of the pitch slope at the end of an utterance. Further, we discuss implications of a corpus-based approach to BC prediction evaluation.

Original language	Undefined
Title of host publication	Proceedings of Interspeech 2010
Publisher	International Speech Communication Association
Pages	3058-3061
Number of pages	4
ISBN (Print)	1990-9772
Publication status	Published - Sept 2010
Event	11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010 - Makuhari, Chiba, Japan Duration: 26 Sept 2010 → 30 Sept 2010 Conference number: 11 http://www.interspeech2010.jpn.org/

Publication series

Name
Publisher	International Speech Communication Association (ISCA)
ISSN (Print)	1990-9772

Conference

Conference	11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010
Abbreviated title	INTERSPEECH
Country/Territory	Japan
City	Makuhari, Chiba
Period	26/09/10 → 30/09/10
Internet address	http://www.interspeech2010.jpn.org/

Keywords

METIS-271083
Backchannel prediction
HMI-SLT: Speech and Language Technology
Prosody
EWI-18627
IR-74048

Access to Document

Cite this

@inproceedings{ab422fd207014e58b5b84012f9f999ef,

title = "A rule-based backchannel prediction model using pitch and pause information",

abstract = "We manually designed rules for a backchannel (BC) prediction model based on pitch and pause information. In short, the model predicts a BC when there is a pause of a certain length that is preceded by a falling or rising pitch. This model was validated against the Dutch IFADV Corpus in a corpus-based evaluation method. The results showed that our model performs slightly better than another well-known rule-based BC prediction model that uses only pitch information. We observed that the length of a pause preceding a BC is one of the important features in this model, next to the duration of the pitch slope at the end of an utterance. Further, we discuss implications of a corpus-based approach to BC prediction evaluation.",

keywords = "METIS-271083, Backchannel prediction, HMI-SLT: Speech and Language Technology, Prosody, EWI-18627, IR-74048",

author = "Truong, {Khiet Phuong} and Poppe, {Ronald Walter} and Heylen, {Dirk K.J.}",

year = "2010",

month = sep,

language = "Undefined",

isbn = "1990-9772",

publisher = "International Speech Communication Association",

pages = "3058--3061",

booktitle = "Proceedings of Interspeech 2010",

address = "France",

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

url = "http://www.interspeech2010.jpn.org/",

}

TY - GEN

T1 - A rule-based backchannel prediction model using pitch and pause information

AU - Truong, Khiet Phuong

AU - Poppe, Ronald Walter

AU - Heylen, Dirk K.J.

N1 - Conference code: 11

PY - 2010/9

Y1 - 2010/9

N2 - We manually designed rules for a backchannel (BC) prediction model based on pitch and pause information. In short, the model predicts a BC when there is a pause of a certain length that is preceded by a falling or rising pitch. This model was validated against the Dutch IFADV Corpus in a corpus-based evaluation method. The results showed that our model performs slightly better than another well-known rule-based BC prediction model that uses only pitch information. We observed that the length of a pause preceding a BC is one of the important features in this model, next to the duration of the pitch slope at the end of an utterance. Further, we discuss implications of a corpus-based approach to BC prediction evaluation.

AB - We manually designed rules for a backchannel (BC) prediction model based on pitch and pause information. In short, the model predicts a BC when there is a pause of a certain length that is preceded by a falling or rising pitch. This model was validated against the Dutch IFADV Corpus in a corpus-based evaluation method. The results showed that our model performs slightly better than another well-known rule-based BC prediction model that uses only pitch information. We observed that the length of a pause preceding a BC is one of the important features in this model, next to the duration of the pitch slope at the end of an utterance. Further, we discuss implications of a corpus-based approach to BC prediction evaluation.

KW - METIS-271083

KW - Backchannel prediction

KW - HMI-SLT: Speech and Language Technology

KW - Prosody

KW - EWI-18627

KW - IR-74048

M3 - Conference contribution

SN - 1990-9772

SP - 3058

EP - 3061

BT - Proceedings of Interspeech 2010

PB - International Speech Communication Association

T2 - 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010

Y2 - 26 September 2010 through 30 September 2010

ER -