Detecting friendly, flirtatious, awkward, and assertive speech in speed-dates

Rajesh Ranganath, Dan Jurafsky, Daniel A. McFarland

Research output: Contribution to journalArticle

Abstract

Automatically detecting human social intentions and attitudes from spoken conversation is an important task for speech processing and social computing. We describe a system for detecting interpersonal stance: whether a speaker is flirtatious, friendly, awkward, or assertive. We make use of a new spoken corpus of over 1000 4-min speed-dates. Participants rated themselves and their interlocutors for these interpersonal stances, allowing us to build detectors for style both as interpreted by the speaker and as perceived by the hearer. We use lexical, prosodic, and dialog features in an SVM classifier to detect very clear styles (the strongest 10 in each stance) with up to 75 accuracy on previously seen speakers (50 baseline) and up to 59 accuracy on new speakers (48 baseline). A feature analysis suggests that flirtation is marked by joint focus on the woman as a target of the conversation, awkwardness by decreased speaker involvement, and friendliness by a conversational style including other-directed laughter and appreciations. Our work has implications for our understanding of interpersonal stance, their linguistic expression, and their automatic extraction.

Original languageEnglish (US)
Pages (from-to)89-115
Number of pages27
JournalComputer Speech and Language
Volume27
Issue number1
DOIs
StatePublished - Jan 1 2013

Fingerprint

Speech processing
Date
Linguistics
Classifiers
Detectors
Baseline
Social Computing
Speech Processing
Classifier
Detector
Target
Speech
Style

Keywords

  • Dating
  • Emotion
  • Paralinguistics
  • Prosody

ASJC Scopus subject areas

  • Software
  • Theoretical Computer Science
  • Human-Computer Interaction

Cite this

Detecting friendly, flirtatious, awkward, and assertive speech in speed-dates. / Ranganath, Rajesh; Jurafsky, Dan; McFarland, Daniel A.

In: Computer Speech and Language, Vol. 27, No. 1, 01.01.2013, p. 89-115.

Research output: Contribution to journalArticle

Ranganath, Rajesh ; Jurafsky, Dan ; McFarland, Daniel A. / Detecting friendly, flirtatious, awkward, and assertive speech in speed-dates. In: Computer Speech and Language. 2013 ; Vol. 27, No. 1. pp. 89-115.
@article{76c2d7287fb24f86bb33f8d16827ac28,
title = "Detecting friendly, flirtatious, awkward, and assertive speech in speed-dates",
abstract = "Automatically detecting human social intentions and attitudes from spoken conversation is an important task for speech processing and social computing. We describe a system for detecting interpersonal stance: whether a speaker is flirtatious, friendly, awkward, or assertive. We make use of a new spoken corpus of over 1000 4-min speed-dates. Participants rated themselves and their interlocutors for these interpersonal stances, allowing us to build detectors for style both as interpreted by the speaker and as perceived by the hearer. We use lexical, prosodic, and dialog features in an SVM classifier to detect very clear styles (the strongest 10 in each stance) with up to 75 accuracy on previously seen speakers (50 baseline) and up to 59 accuracy on new speakers (48 baseline). A feature analysis suggests that flirtation is marked by joint focus on the woman as a target of the conversation, awkwardness by decreased speaker involvement, and friendliness by a conversational style including other-directed laughter and appreciations. Our work has implications for our understanding of interpersonal stance, their linguistic expression, and their automatic extraction.",
keywords = "Dating, Emotion, Paralinguistics, Prosody",
author = "Rajesh Ranganath and Dan Jurafsky and McFarland, {Daniel A.}",
year = "2013",
month = "1",
day = "1",
doi = "10.1016/j.csl.2012.01.005",
language = "English (US)",
volume = "27",
pages = "89--115",
journal = "Computer Speech and Language",
issn = "0885-2308",
publisher = "Academic Press Inc.",
number = "1",

}

TY - JOUR

T1 - Detecting friendly, flirtatious, awkward, and assertive speech in speed-dates

AU - Ranganath, Rajesh

AU - Jurafsky, Dan

AU - McFarland, Daniel A.

PY - 2013/1/1

Y1 - 2013/1/1

N2 - Automatically detecting human social intentions and attitudes from spoken conversation is an important task for speech processing and social computing. We describe a system for detecting interpersonal stance: whether a speaker is flirtatious, friendly, awkward, or assertive. We make use of a new spoken corpus of over 1000 4-min speed-dates. Participants rated themselves and their interlocutors for these interpersonal stances, allowing us to build detectors for style both as interpreted by the speaker and as perceived by the hearer. We use lexical, prosodic, and dialog features in an SVM classifier to detect very clear styles (the strongest 10 in each stance) with up to 75 accuracy on previously seen speakers (50 baseline) and up to 59 accuracy on new speakers (48 baseline). A feature analysis suggests that flirtation is marked by joint focus on the woman as a target of the conversation, awkwardness by decreased speaker involvement, and friendliness by a conversational style including other-directed laughter and appreciations. Our work has implications for our understanding of interpersonal stance, their linguistic expression, and their automatic extraction.

AB - Automatically detecting human social intentions and attitudes from spoken conversation is an important task for speech processing and social computing. We describe a system for detecting interpersonal stance: whether a speaker is flirtatious, friendly, awkward, or assertive. We make use of a new spoken corpus of over 1000 4-min speed-dates. Participants rated themselves and their interlocutors for these interpersonal stances, allowing us to build detectors for style both as interpreted by the speaker and as perceived by the hearer. We use lexical, prosodic, and dialog features in an SVM classifier to detect very clear styles (the strongest 10 in each stance) with up to 75 accuracy on previously seen speakers (50 baseline) and up to 59 accuracy on new speakers (48 baseline). A feature analysis suggests that flirtation is marked by joint focus on the woman as a target of the conversation, awkwardness by decreased speaker involvement, and friendliness by a conversational style including other-directed laughter and appreciations. Our work has implications for our understanding of interpersonal stance, their linguistic expression, and their automatic extraction.

KW - Dating

KW - Emotion

KW - Paralinguistics

KW - Prosody

UR - http://www.scopus.com/inward/record.url?scp=84867332819&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84867332819&partnerID=8YFLogxK

U2 - 10.1016/j.csl.2012.01.005

DO - 10.1016/j.csl.2012.01.005

M3 - Article

VL - 27

SP - 89

EP - 115

JO - Computer Speech and Language

JF - Computer Speech and Language

SN - 0885-2308

IS - 1

ER -