logging in or signing up TTS personality and choice Roxie Download Post to : URL : Related Presentations : Share Add to Flag Embed Email Send to Blogs and Networks Add to Channel Uploaded from authorPOINT Insert YouTube videos in PowerPont slides with aS Desktop Copy embed code: (To copy code, click on the text box) Embed: URL: Thumbnail: WordPress Embed Customize Embed The presentation is successfully added In Your Favorites. Views: 246 Category: News & Reports.. License: All Rights Reserved Like it (0) Dislike it (0) Added: August 11, 2007 This Presentation is Public Favorites: 0 Presentation Description No description available. Comments Posting comment... Premium member Presentation Transcript Slide1: TTS Personality, Similarity-Attraction, and Choice Kwan M. Lee Key Questions: Key Questions Can computer-generated speech manifest personality? How important is the manifestation of personality? Importance of the Questions: Importance of the Questions Psychological Issues Human voice manifests personality Could clearly non-human TTS do the same? Less primitive than gender Does similarity-attraction work with voices? Is there proximate source orientation? Importance of the Questions: Importance of the Questions Design Issues TTS parameters are readily selected How should we choose the settings? Context: Context Book purchase site on the web Five book reviews, each followed by questions Experimental Design: Experimental Design IV1: Personality of subject Myers-Briggs Personality scale Wiggins Introvert adjective index Wiggins Extrovert adjective index IV2: Coding of subject’s voice for introversion/extroversion Manipulation of TTS personality: Manipulation of TTS personality CSLU Toolkit for TTS Extrovert voice F0: 140Hz (F0) Pitch range: 40Hz Words per minute: 181 Volume level: 100% Introvert voice F0: 82 Hz (F0) Pitch range: 16 Hz Words per minute: 154 Volume level: 15% Extroverted Voice: Extroverted Voice Introverted Voice: Introverted Voice Extroversion of TTS Voice: Extroversion of TTS Voice Buying Intention: Buying Intention Quality of the Book Review: Quality of the Book Review Credibility of the Reviewer: Credibility of the Reviewer Other Results: Other Results Reliability of subject’s voice coding = .8 Correlation of subject’s voice and subject’s personality = .23 (signif., but small) Similarity of subject’s voice and TTS voice does not have any effect Psychological Implications: Psychological Implications TTS voice manifests personality Voices are social actors Even when obviously non-human Similarity-attraction operates with subject’s personality (not person’s voice) and TTS personality Buying intention Perceived quality Credibility Personality of voice has effects beyond the proximate Reviewer Design Implications: Design Implications Personality types are easily created Personality is powerful Personality is a key individual difference Personality does not require AI 'Casting' of TTS personality Follow-up with Choice: Follow-up with Choice Key Questions: Key Questions Will users choose TTS voice according to their own personality? Does computer generated speech manifest personality even when users have a choice? Does choice affect previous results? Importance of the Questions: Importance of the Questions Psychological Issues Assignment vs. Choice: The similarity-attraction experiments upside down Cognitive dissonance vs. Similarity attraction Predicting personality from choice behaviors Importance of the Questions: Importance of the Questions Design Issues Empirical test of the value of choice in HCI Do people choose matching personality?: Do people choose matching personality? Classify participants as extrovert or introvert Wiggins extrovert scale Wiggins introvert scale Myers-Briggs E/I scale Participants do not choose a TTS voice according to his/her personality. Does TTS manifest personality?: Does TTS manifest personality? Is there proximate source orientation? : Is there proximate source orientation? Is there similarity-attraction? : Is there similarity-attraction? No significant similarity attraction effects voice attractiveness voice credibility quality of the review liking of the reviewer credibility of the reviewer user’s buying intention Choice is the key predictor Cognitive dissonance overwhelms similarity-attraction Psychological Implications: Psychological Implications TTS voice manifests personality even when chosen by a user Proximate source orientation is very powerful Even when voice is obviously arbitrary! Cognitive dissonance weakens similarity-attraction Initial choice determines later social responses Design Implications: Design Implications People automatically assign a personality to TTS voices Personality of voice affects all subsequent perceptions Casting issues Choice matters!: Commitment leads to positive social responses You do not have the permission to view this presentation. In order to view it, please contact the author of the presentation.
TTS personality and choice Roxie Download Post to : URL : Related Presentations : Share Add to Flag Embed Email Send to Blogs and Networks Add to Channel Uploaded from authorPOINT Insert YouTube videos in PowerPont slides with aS Desktop Copy embed code: (To copy code, click on the text box) Embed: URL: Thumbnail: WordPress Embed Customize Embed The presentation is successfully added In Your Favorites. Views: 246 Category: News & Reports.. License: All Rights Reserved Like it (0) Dislike it (0) Added: August 11, 2007 This Presentation is Public Favorites: 0 Presentation Description No description available. Comments Posting comment... Premium member Presentation Transcript Slide1: TTS Personality, Similarity-Attraction, and Choice Kwan M. Lee Key Questions: Key Questions Can computer-generated speech manifest personality? How important is the manifestation of personality? Importance of the Questions: Importance of the Questions Psychological Issues Human voice manifests personality Could clearly non-human TTS do the same? Less primitive than gender Does similarity-attraction work with voices? Is there proximate source orientation? Importance of the Questions: Importance of the Questions Design Issues TTS parameters are readily selected How should we choose the settings? Context: Context Book purchase site on the web Five book reviews, each followed by questions Experimental Design: Experimental Design IV1: Personality of subject Myers-Briggs Personality scale Wiggins Introvert adjective index Wiggins Extrovert adjective index IV2: Coding of subject’s voice for introversion/extroversion Manipulation of TTS personality: Manipulation of TTS personality CSLU Toolkit for TTS Extrovert voice F0: 140Hz (F0) Pitch range: 40Hz Words per minute: 181 Volume level: 100% Introvert voice F0: 82 Hz (F0) Pitch range: 16 Hz Words per minute: 154 Volume level: 15% Extroverted Voice: Extroverted Voice Introverted Voice: Introverted Voice Extroversion of TTS Voice: Extroversion of TTS Voice Buying Intention: Buying Intention Quality of the Book Review: Quality of the Book Review Credibility of the Reviewer: Credibility of the Reviewer Other Results: Other Results Reliability of subject’s voice coding = .8 Correlation of subject’s voice and subject’s personality = .23 (signif., but small) Similarity of subject’s voice and TTS voice does not have any effect Psychological Implications: Psychological Implications TTS voice manifests personality Voices are social actors Even when obviously non-human Similarity-attraction operates with subject’s personality (not person’s voice) and TTS personality Buying intention Perceived quality Credibility Personality of voice has effects beyond the proximate Reviewer Design Implications: Design Implications Personality types are easily created Personality is powerful Personality is a key individual difference Personality does not require AI 'Casting' of TTS personality Follow-up with Choice: Follow-up with Choice Key Questions: Key Questions Will users choose TTS voice according to their own personality? Does computer generated speech manifest personality even when users have a choice? Does choice affect previous results? Importance of the Questions: Importance of the Questions Psychological Issues Assignment vs. Choice: The similarity-attraction experiments upside down Cognitive dissonance vs. Similarity attraction Predicting personality from choice behaviors Importance of the Questions: Importance of the Questions Design Issues Empirical test of the value of choice in HCI Do people choose matching personality?: Do people choose matching personality? Classify participants as extrovert or introvert Wiggins extrovert scale Wiggins introvert scale Myers-Briggs E/I scale Participants do not choose a TTS voice according to his/her personality. Does TTS manifest personality?: Does TTS manifest personality? Is there proximate source orientation? : Is there proximate source orientation? Is there similarity-attraction? : Is there similarity-attraction? No significant similarity attraction effects voice attractiveness voice credibility quality of the review liking of the reviewer credibility of the reviewer user’s buying intention Choice is the key predictor Cognitive dissonance overwhelms similarity-attraction Psychological Implications: Psychological Implications TTS voice manifests personality even when chosen by a user Proximate source orientation is very powerful Even when voice is obviously arbitrary! Cognitive dissonance weakens similarity-attraction Initial choice determines later social responses Design Implications: Design Implications People automatically assign a personality to TTS voices Personality of voice affects all subsequent perceptions Casting issues Choice matters!: Commitment leads to positive social responses