A5029071832- AI in Service Interactions
- Speech and dialogue systems
- Language, Discourse, Communication Strategies
- Social Robot Interaction and HRI
- Action Observation and Synchronization
- Reinforcement Learning in Robotics
- Neurobiology of Language and Bilingualism
- Advanced Decision-Making Techniques
University of Electronic Science and Technology of China
2024
University College Dublin
2020-2023
Limited linguistic coverage for Intelligent Personal Assistants (IPAs) means that many interact in a non-native language. Yet we know little about how IPAs currently support or hinder these users. Through native (L1) and (L2) English speakers interacting with Google Assistant on smartphone smart speaker, aim to understand this more deeply. Interviews revealed L2 prioritised utterance planning around perceived limitations, as opposed L1 prioritising succinctness because of system limitations....
Through smartphones and smart speakers, intelligent personal assistants (IPAs) have made speech a common interaction modality. With linguistic coverage varying functionality levels, many speakers engage with IPAs using non-native language. This may impact mental workload patterns of language production used by speakers. We present mixed-design experiment, where native (L1) (L2) English completed tasks via found significantly higher for L2 in IPA interactions. Contrary to our hypotheses, we...
Intelligent Personal Assistants (IPAs) are limited in the languages they support, meaning many people left to interact using a non-native language. Yet, we know little about how with IPAs this way. Through conversation analysis (CA) perspective, examine native (L1) and (L2) English speaker interactions Google Assistant, comparing both user groups produce IPA commands. Our work shows that L1 L2 speakers similarly used pauses, partial or complete repetition, hyper-articulation when...
Human-machine dialogue (HMD) research debates the degree to which language production in this context is egocentric or allocentric. That is, a person might take machine's perspective into account. Our study aims identify whether users produce allocentric within speech-based HMD when there asymmetry information available both partners. Through an adapted referential communication task, we manipulated presence absence of visual distractors and occlusions, similarly previous tasks used...
Conversational User Interfaces such as Voice Assistants are hugely popular. Yet they designed to be monolingual by default, lacking support for, or sensitivity to, the bilingual dialogue experience. In this provocation paper, we highlight language production challenges faced in VA interaction for users. We argue that, facilitating phenomena seen interaction, code-switching, can foster a more inclusive and improved user experience also explore ways that might achieved, through of multiple...