On the Potential of Modular Voice Conversion for Virtual Agents

Mertes, Silvan; Kiderle, Thomas; Schlagowski, Ruben; Lingenfelser, Florian; André, Elisabeth

doi:10.1109/aciiw52867.2021.9666349

Cited by 4 publications

(2 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, a mismatched voice can result in a variety of aversive reactions, such as unsettling, eerie, uncanny, and repulsive responses [51,54,55,74,78]. The intensity of this dissonance can be influenced by factors like the user's age or the robot's realism [17,52].…”

Section: Introductionmentioning

confidence: 99%

Giving Robots a Voice: Human-in-the-Loop Voice Creation and open-ended Labeling

van Rijn,

Mertes,

Janowski

et al. 2024

Proceedings of the CHI Conference on Human Factors in Computing Systems

Self Cite

View full text Add to dashboard Cite

Section: Introductionmentioning

confidence: 99%

Giving Robots a Voice: Human-in-the-Loop Voice Creation and open-ended Labeling

van Rijn,

Mertes,

Janowski

et al. 2024

Proceedings of the CHI Conference on Human Factors in Computing Systems

Self Cite

View full text Add to dashboard Cite

“…However, for many realworld use cases, such as emotional virtual agents, corresponding face images need to be generated in a more detailed manner to improve the credibility and anthropomorphism of human-like avatars. This is especially of interest during the design stage of such virtual agents, as the consistency of different modalities of virtual agents is of great importance (Gong and Nass, 2007;Mertes et al, 2021a) and fine-grained degrees of expressivity can enhance the perception of certain affective states of the agent, influencing (among others) the perception of the agent's personality (Kiderle et al, 2021).…”

Section: Introductionmentioning

confidence: 99%

Intercategorical Label Interpolation for Emotional Face Generation with Conditional Generative Adversarial Networks

Mertes¹,

Schiller²,

Lingenfelser³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Generative adversarial networks offer the possibility to generate deceptively real images that are almost indistinguishable from actual photographs. Such systems however rely on the presence of large datasets to realistically replicate the corresponding domain. This is especially a problem if not only random new images are to be generated, but specific (continuous) features are to be co-modeled. A particularly important use case in Human-Computer Interaction (HCI) research is the generation of emotional images of human faces, which can be used for various use cases, such as the automatic generation of avatars. The problem hereby lies in the availability of training data. Most suitable datasets for this task rely on categorical emotion models and therefore feature only discrete annotation labels. This greatly hinders the learning and modeling of smooth transitions between displayed affective states. To overcome this challenge, we explore the potential of label interpolation to enhance networks trained on categorical datasets with the ability to generate images conditioned on continuous features.

show abstract