Towards measuring continuous acoustic feature convergence in unconstrained spoken dialogues

Spyros Kousidis, David Dorran, Yi Wang, Brian Vaughan, Charlie Cullen, Dermot Campbell, Ciaran McDonnell, Eugene Coyle

Research output: Contribution to conferencePaper

Abstract

Acoustic/prosodic feature (a/p) convergence has been known to occur both in dialogues between humans, as well as in human-computer interactions. Understanding the form and function of convergence is desirable for developing next generation conversational agents, as this will help increase speech recognition performance and naturalness of synthesized speech. Currently, the underlying mechanisms by which continuous and bi-directional convergence occurs are not well understood. In this study, a direct comparison between time-aligned frames shows significant similarity in acoustic feature variation between the two speakers. The method described (TAMA) constitutes a first step towards a quantitative analysis of a/p convergence.
Original languageEnglish
Publication statusPublished - 22 Sep 2008
Externally publishedYes
EventInterspeech 2008 - Brisbane, Australia
Duration: 22 Sep 200826 Sep 2008

Conference

ConferenceInterspeech 2008
CountryAustralia
CityBrisbane
Period22/09/0826/09/08

Fingerprint

Acoustics
Human computer interaction
Speech recognition
Chemical analysis

Cite this

Kousidis, S., Dorran, D., Wang, Y., Vaughan, B., Cullen, C., Campbell, D., ... Coyle, E. (2008). Towards measuring continuous acoustic feature convergence in unconstrained spoken dialogues. Paper presented at Interspeech 2008, Brisbane, Australia.
Kousidis, Spyros ; Dorran, David ; Wang, Yi ; Vaughan, Brian ; Cullen, Charlie ; Campbell, Dermot ; McDonnell, Ciaran ; Coyle, Eugene. / Towards measuring continuous acoustic feature convergence in unconstrained spoken dialogues. Paper presented at Interspeech 2008, Brisbane, Australia.
@conference{8e02d20270274a7cb5a231106f522205,
title = "Towards measuring continuous acoustic feature convergence in unconstrained spoken dialogues",
abstract = "Acoustic/prosodic feature (a/p) convergence has been known to occur both in dialogues between humans, as well as in human-computer interactions. Understanding the form and function of convergence is desirable for developing next generation conversational agents, as this will help increase speech recognition performance and naturalness of synthesized speech. Currently, the underlying mechanisms by which continuous and bi-directional convergence occurs are not well understood. In this study, a direct comparison between time-aligned frames shows significant similarity in acoustic feature variation between the two speakers. The method described (TAMA) constitutes a first step towards a quantitative analysis of a/p convergence.",
author = "Spyros Kousidis and David Dorran and Yi Wang and Brian Vaughan and Charlie Cullen and Dermot Campbell and Ciaran McDonnell and Eugene Coyle",
year = "2008",
month = "9",
day = "22",
language = "English",
note = "Interspeech 2008 ; Conference date: 22-09-2008 Through 26-09-2008",

}

Kousidis, S, Dorran, D, Wang, Y, Vaughan, B, Cullen, C, Campbell, D, McDonnell, C & Coyle, E 2008, 'Towards measuring continuous acoustic feature convergence in unconstrained spoken dialogues' Paper presented at Interspeech 2008, Brisbane, Australia, 22/09/08 - 26/09/08, .

Towards measuring continuous acoustic feature convergence in unconstrained spoken dialogues. / Kousidis, Spyros; Dorran, David; Wang, Yi; Vaughan, Brian; Cullen, Charlie; Campbell, Dermot; McDonnell, Ciaran; Coyle, Eugene.

2008. Paper presented at Interspeech 2008, Brisbane, Australia.

Research output: Contribution to conferencePaper

TY - CONF

T1 - Towards measuring continuous acoustic feature convergence in unconstrained spoken dialogues

AU - Kousidis, Spyros

AU - Dorran, David

AU - Wang, Yi

AU - Vaughan, Brian

AU - Cullen, Charlie

AU - Campbell, Dermot

AU - McDonnell, Ciaran

AU - Coyle, Eugene

PY - 2008/9/22

Y1 - 2008/9/22

N2 - Acoustic/prosodic feature (a/p) convergence has been known to occur both in dialogues between humans, as well as in human-computer interactions. Understanding the form and function of convergence is desirable for developing next generation conversational agents, as this will help increase speech recognition performance and naturalness of synthesized speech. Currently, the underlying mechanisms by which continuous and bi-directional convergence occurs are not well understood. In this study, a direct comparison between time-aligned frames shows significant similarity in acoustic feature variation between the two speakers. The method described (TAMA) constitutes a first step towards a quantitative analysis of a/p convergence.

AB - Acoustic/prosodic feature (a/p) convergence has been known to occur both in dialogues between humans, as well as in human-computer interactions. Understanding the form and function of convergence is desirable for developing next generation conversational agents, as this will help increase speech recognition performance and naturalness of synthesized speech. Currently, the underlying mechanisms by which continuous and bi-directional convergence occurs are not well understood. In this study, a direct comparison between time-aligned frames shows significant similarity in acoustic feature variation between the two speakers. The method described (TAMA) constitutes a first step towards a quantitative analysis of a/p convergence.

UR - https://www.semanticscholar.org/paper/Towards-measuring-continuous-acoustic-feature-in-Kousidis-Dorran/59a88c3532471876cf441d63d7a168321dbda626

M3 - Paper

ER -

Kousidis S, Dorran D, Wang Y, Vaughan B, Cullen C, Campbell D et al. Towards measuring continuous acoustic feature convergence in unconstrained spoken dialogues. 2008. Paper presented at Interspeech 2008, Brisbane, Australia.