Recent work has looked to understand user perceptions of speech agent capabilities as dialogue partners (termed partner models), and how this affects user interaction. Yet, partner model effects are currently inferred from language production as no metrics are available to quantify these subjective perceptions more directly. Through three phases of work, we develop and validate the Partner Modelling Questionnaire (PMQ): an 18-item self-report semantic differential scale designed to reliably measure people’s partner models of non-embodied speech interfaces. Through confirmatory factor analysis, we confirm that the PMQ scale consists of three factors: communicative competence and dependability, human-likeness in communication and communicative flexibility. Our studies show that the measure consistently demonstrates good internal reliability, strong test-retest reliability over 4- and 12-week intervals, and predictable convergent/divergent validity. Based on our findings, we discuss the multidimensional nature of partner models, while identifying key future research avenues that the development of the PMQ facilitates. Notably, this includes the need to identify the activation, sensitivity, and dynamism of partner models in speech interface interaction.

Gestational diabetes mellitus (GDM) is an increasingly prominent health issue in pregnant women. While various technology solutions have been developed to support self-management of women with GDM, usability and functionality limitations have precluded their adoption. More active engagement with women with GDM in the design process could mitigate these limitations, thus we developed a design method to support the involvement of women with GDM in the design phase of a GDM self-management system. Thirteen online workshops were conducted involving five women with GDM and two postpartum women with previous GDM, who participated in idea generation, paper-based sketching, and group discussions, followed by interviews to gather their experiences of participation. We found that women valued their inclusion in these design workshops and felt confident sharing their ideas, from which we introduce recommendations for design procedures that enhance the contributions of women with GDM in the design of self-management apps.

Speech-enabled smart speakers are common devices used for numerous tasks in everyday life. While speech-enabled technologies are widespread, using one’s voice as a computing modality introduces new accessibility challenges for people with speech disfluencies such as stammering (also known as stuttering). This paper investigates the smart speaker user experiences of people who stammer over three weeks. We conducted diary studies and semi-structured interviews with 11 individuals to identify their daily routines, difficulties with successful interactions, and strategies to overcome these barriers. Our analysis demonstrates key factors such as device location, its affordances, and the structure of commands had a strong impact on user experience. Participants highlighted different linguistic strategies to try and overcome interaction difficulties and discussed the potential of using smart speakers for speech and language therapy. We emphasise the need to further understand the experiences of people who stammer in smart speaker design to increase their accessibility.

Casual conversation, where people engage in social talk or phatic communion (Coupland et al., 1992) (i.e., non-task oriented talk) is acknowledged as important in facilitating collaboration among c...

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 non-native (L2) English speakers interacting with Google Assistant on a smartphone and smart speaker, we aim to understand this more deeply. Interviews revealed that L2 speakers prioritised utterance planning around perceived linguistic limitations, as opposed to L1 speakers prioritising succinctness because of system limitations. L2 speakers see IPAs as insensitive to linguistic needs resulting in failed interaction. L2 speakers clearly preferred using smartphones, as visual feedback supported diagnoses of communication breakdowns whilst allowing time to process query results. Conversely, L1 speakers preferred smart speakers, with audio feedback being seen as sufficient. We discuss the need to tailor the IPA experience for L2 users, emphasising visual feedback whilst reducing the burden of language production.

Integrated into smartphones or smart speakers, Intelligent Personal Assistants (IPAs) have grown into the most prevalent example of speech-based interfaces today. Enabling hands-free completion of tasks such as checking the weather, playing music, or controlling smart devices, IPAs have become a central feature of millions of homes. Yet, end-users are faced with barriers to understanding their `black box' devices, and a lack of opportunity to engage in end-user development (EUD) activities. Our current work considers the state of existing tools and platforms for development of IPAs, the key challenges to be overcome, and the potential benefits of doing so.

Psychology researchers employ the Experience Sampling Method (ESM) to capture thoughts and behaviours of participants within their everyday lives. Smartphone-based ESM apps are increasingly used in such research. However, the diversity of researchers' app requirements, coupled with cost and complexity of their implementation, has prompted end-user development (EUD) approaches. In addition, limited evaluation of such environments beyond lab-based usability studies precludes discovery of factors pertaining to real-world EUD adoption. We first describe the extension of Jeeves, our visual programming environment for ESM app creation, in which we implemented additional functional requirements, derived from a survey and analysis of previous work. We further describe interviews with psychology researchers to understand their practical considerations for employing this extended environment in their work practices. Results of our analysis are presented as factors pertaining to the adoption of EUD activities within and between communities of practice.

The Experience Sampling Method (ESM) captures participants' thoughts and feelings in their everyday environments. Mobile and wearable technologies afford us opportunities to reach people using ESM in varying contexts. However, a lack of programming knowledge often hinders researchers in creating ESM applications. In practice, they rely on specialised tools for app creation. Our initial review of these tools indicates that most are expensive commercial services, and none utilise the full potential of sensors for creating context-aware applications. We present “Jeeves”, a visual language to facilitate ESM application creation. Inspired by successful visual languages in literature, our block-based notation enables researchers to visually construct ESM study specifications. We demonstrate its applicability by replicating existing ESM studies found in medical and psychology literature. Our preliminary study with 20 participants demonstrates that both non-programmers and programmers are able to successfully utilise Jeeves. We discuss future work in extending Jeeves with alternative mobile technologies.

Traditional behaviour change interventions are typically delivered with a fixed set of components, providing identical content to all participants in a trial. The disregard of personal differences often leads to weak effects and inconclusive results. Tools are required that let researchers identify effective components for specific users and contexts. This paper presents a system design incorporating user models and a visual programming language to allow end-users with varying technical expertise to develop tailored interventions using feedback from a series of visual and non-visual interfaces.

Dasher is a promising fast assistive gaze communication method. However, previous evaluations of Dasher have been inconclusive. Either the studies have been too short, involved too few participants, suffered from sampling bias, lacked a control condition, used an inappropriate language model, or a combination of the above. To rectify this, we report results from two new evaluations of Dasher carried out using a Tobii P10 assistive eye-tracker machine. We also present a method of modifying Dasher so that it can use a state-of-the-art long-span statistical language model. Our experimental results show that compared to a baseline eye-typing method, Dasher resulted in significantly faster entry rates (12.6 wpm versus 6.0 wpm in Experiment 1, and 14.2 wpm versus 7.0 wpm in Experiment 2). These faster entry rates were possible while maintaining error rates comparable to the baseline eye-typing method. Participants' perceived physical demand, mental demand, effort and frustration were all significantly lower for Dasher. Finally, participants significantly rated Dasher as being more likeable, requiring less concentration and being more fun.