Modelling personality in voices of talking products through prosodic parameters

Abstract

In this paper we report preliminary findings from two user studies that on the one hand investigate how prosodic parameters of synthetic speech can influence the perceived impression of the speakers personality and on the other hand explores if and how people attribute personality to objects such as typical products of daily shopping. The results show that a) prosodic parameters have a strong influence on the perceived personality and can be partially used to achieve a desired impression and b) that subjects clearly attribute personalities to products. Both findings encourage us to continue our work on a dialogue shell for talking products.

A Survey on perceived speaker traits: Personality, likability, pathology, and the first challenge

“…Therefore, speech data should allow one to perform APP reasonably well. While still being limited, the results proposed so far in the speech literature seem to confirm the indications above for both APR (Mairesse et al, 2007;Ivanov et al, 2011) and APP (Mairesse et al, 2007;Mohammadi and Vinciarelli, 2012;Polzehl et al, 2010;Valente et al, 2012;Nass and Min Lee, 2001;Schmitz et al, 2007;Trouvain et al, 2006).…”

“…Such an effect, known as similarity-attraction, takes place between humans as well ) and seems to suggest that people do not make differences, at an unconscious level, between real and artificial voices. In the same vein, the experiments proposed by Schmitz et al (2007) and Trouvain et al (2006) show that it is possible to elicit the attribution of predefined traits by changing the characteristics of artificial voices (36 assessors were involved in the tests).…”

“…The APP studies presented so far deal with real voices, but the literature shows that human listeners tend to assign personality traits to artificial speakers as well (Nass and Min Lee, 2001;Schmitz et al, 2007;Trouvain et al, 2006). Nass and Min Lee (2001) pioneered the manipulation of prosody (intensity, pitch, pitch range, and speaking rate) aiming at producing speech that sounds more or less extroverted.…”

The Shopping Experience of Tomorrow: Human-Centered and Resource-Adaptive

“…We changed the four prosodic parameters pitch range (in semitones), pitch level (in Hz), tempo (as a durational factor in ms), and intensity (soft, modal, or loud as in [57]) according to our literature review [55]. For example, a competent voice has a higher pitch range (8 semitones), a lower pitch level (-30%), a 30% higher tempo, and a loud intesity compared to the baseline voice.…”

“…For example, a competent voice has a higher pitch range (8 semitones), a lower pitch level (-30%), a 30% higher tempo, and a loud intesity compared to the baseline voice. In [55] we also evaluated whether it is possible to model different personalities with the same voice by adjusting these prosodic parameters, such that listeners will recognize the intended personality dimension. The study has shown that there are clear preferences for our prosody-modeled speech synthesis for certain brand personality dimensions.…”