top of page


A voice user interface for Alzheimer's caregivers



New studies predict the number of people living with Alzheimer's disease will triple to 153 million diagnosed cases by 2050. The socio-economic impact of brain diseases will continue to put a burden on society, specially on family members acting as care givers.


  1. How might we reduce caregivers pain points in their daily care duties?

  2. How might we tackle their sense of isolation and depression?

  3. How might we help them get prepared for unexpected situations?


Along with the engineering team from I developed a VUI prototype for the Google Home Assistant which uses a custom relevance model scrapper to provide essential palliative care information to caregivers without the need to interrupt chores and constant task switch.

Key responsibilities:

• quanti/quali research
• ideation
• product design
• lean ux
• voice interface design
• branding
My role in the project

My role in this product design project, which was a part of my UX certificate program at General Assembly, involved leading an end-to-end process.


With a focus on innovation, I eagerly delved into exploring multi-modal user interfaces (voice interaction and graphic UI) as a key aspect of the final product, aiming to enhance user experiences through seamless integration across various interaction modes. 

Stakeholders and collaborators

Throughout the project, I actively collaborated with colleagues to gather feedback, and get various perspectives. I conducted interviews with end users and experts from Alzheimer's Research UK, and gained valuable insights to inform the design process. Additionally, working closely with a machine learning engineer from Visyion360, an XR company owned by Mediapro, I successfully integrated their expertise to build a prototype that showcased the potential impact of our multi-modal user interface concept.


Design Thinking & Human-Centred Design


I utilized design thinking and human-centered design principles for several reasons. Design thinking allowed me to deeply understand users' needs, challenges, and aspirations, fostering empathy throughout the design process. By prioritising the user's perspective, I ensured that the solutions developed directly addressed their specific requirements and goals.


Human-centred design principles promoted iterative and collaborative approaches, facilitating co-creation and feedback from end users, caregivers, and experts. In a care tech project, prioritising human-centered design ensures technology that supports and enhances individuals' well-being, promoting empowerment, independence, and dignity.



The task of caring for a person with Dementia and Alzheimer's often falls upon family members who have to act as caregivers. Caregiving is full of challenges and results in a financial and emotional impact on carers as the condition of a loved one evolves. This often leads to physical and mental fatigue, social isolation and even depression.

According to a study by The Economist Intelligence Unit, dementia can cost a patient an average of £40.7k per year in the UK. Other recent studies point to a global total of 155million cases by 2050. It is estimated that in the UK, a caregiver is on average £18.2K worse off as a result of lost income, productivity and opportunities.

Market need

My initial research on the market landscape showed a number of supporting services that are focused on the dementia patients. I believe that not enough help is available to caregivers themselves.


The tools and services such as assistive tech can be fragmented and provide poor interoperability with social, communication and productivity platforms among others, which limits their utility and impact. Therefore, this is a domain that is very much in need of innovative solutions and provide a worthwhile user-centric challenge. 

Report by The Economist Intelligence Unit on the impact of Alzheimer's on the UK economy


Contextual inquiry

Desk research


For the desk research exercise in the area of Alzheimer's and Dementia caregiving, I extensively reviewed reports from credible sources such as Alzheimer's Society UK and Alzheimer's Research UK. The Economist Intelligence Unit report on the subject was also used as research material, providing valuable insights from a socio-economic perspective.


Furthermore, various articles and sources from academic journals, healthcare publications, and caregiver forums were explored, ensuring a comprehensive understanding of the challenges and existing solutions in this field.

Ethnographic research:

qualitative interviews


I conducted six interviews over the phone with family caregivers to understand their context and challenges, taking into consideration the varying stages in the condition and the different levels of care required by their loved ones.


Additionally, I interviewed two active senior care home managers responsible for all aspects of operations, which gave me a utilitarian perspective on the caregiving process from a professional point of view.

Quantitative survey

Finally, I carried out a quantitative survey with users in the UK, Germany and Spain, to confirm/disprove some of the patterns that emerged during the interviews. I also got more data about questions related to the use of technology by carers.

At the end of the survey, I encouraged respondents to describe their daily activity in the form of a summarised diary. That provided valuable information into user's pain points, blockers and more when carrying out their daily activities.

The consultation with users had not stopped there though. Throughout the development process there had been one-off talks with different caregivers in the three surveyed regions, which was valuable in terms of quick feedback loops.

2 Online Survey.jpg

Defining the problem space

Space saturation and affinity mapping

Based on the research during the Empathy phase I organised the collected data into an information wall, gradually shaping it  as an affinity map.


I involved other UX designers in the process to help me map out the more dense data sets. That included findings in the primary immersion, user interviews and online survey.

The second round received the contribution of a creative technology strategist who helped consolidate the affinity map into workable categories: 'demographics'; 'problems/issues/challenges/pain-points'; 'needs'; and 'habits and behaviour'.

Identifying patterns and scoping the problem space


From the affinity map I identified three recurrent critical areas impacting caregivers: preparedness, support, and isolation.

Because of the evident frequency with which these three themes appeared in all interviews, I decided that they would form the pillars of the problem space.

Salient pain points


As a caregiver I need relevant information about what the progression and impact of Alzheimer's on my loved one just so I can plan things in advance


As a caregiver I need all the support I can get from family, friends, and technology. It would be impossible to this on my own


As a caregiver I need social interaction and emotional connection, especially with others in the same position, just so I can share, learn, feel less lonely and more supported

Goal-oriented persona

The proto-caregiver

The caregiving scenarios for dementia patients can vary significantly. The data I collected indicated that one of the most common situations involves adult offspring caring for their elderly parents, while spouses often take care of their partners.


The group and gender distribution can be diverse, but during my research, I observed that the majority of cases involved female carers, either spouses caring for their husbands or female offspring looking after a parent.


To create a representative primary persona, I selected a middle-aged female caregiver who cared for her husband, as this group appeared slightly more prevalent within the surveyed sample.


Family caregivers need a way to get continuous support throughout their caring journey for their loved ones because the demands are very real and can lead to self-neglect, isolation, and physical and mental exhaustion.


"I believe that by building a multi-modal assistive system centred on voice interaction that provides guidance for family caregivers of loved ones with Alzheimer's  I will reduce friction during multi-tasking, give valuable information, and ease their sense of isolation and burnout."

Product ideation

User journey storyboards


I revisited the interviews and user diaries to construct a more comprehensive caregiving journey throughout the day. A crucial focus was identifying the key moments when caregivers encountered barriers or experienced friction while performing their tasks.


As my hypothesis was based on a hands-free, conversational interface solution I envisioned three initial use cases to employ that technology: finding relevant information on the spot without stopping a task; connecting with a network of helpers/supporters; and make hands-free calls.

Ideation methods


When teaming up with other UX designers I facilitated and contributed to the ideation session. My technique of choice was the 'Crazy 8's' followed by 'Build, Save, Kill' and, finally, 'dot-voting' to select the best solution.

Nuria_Use Cases_V1-1.jpg
Storyboard 1.jpg


VUI dialogue sample


The next step consisted of prototyping the voice agent, beginning with a well-defined dialogue flow for the ideal user interaction. I decided to develop it from scratch to have full control over its critical attributes: tone, language/vocabulary, voice talent, and fulfilment thresholds.


The development considered the three primary use cases: onboarding, finding information, and connecting with friends, and seeking support

Series of dialogue flows based on the use cases covered by the prototype
Graphical User Interface (GUI) user flow


By prioritising voice as the primary means of interaction, a graphical user interface (GUI) was introduced as a complementary feature to address situations where sound may not be practical or effective. Considering the nature of these interactions as multiple micro-moments, with the majority but not all being voice-based, a multi-modal system was deemed necessary to accommodate various scenarios. Some interactions were more suited for voice, while others were better suited for a screen.


Designing for both required careful consideration of surface switch capabilities. For instance, the onboarding process primarily utilised a smartphone device, but it often involved starting with voice and transitioning to the screen.

Usability testing

Guerrilla testing

To quickly validate my assumptions about the core functions of the voice agent, I translated them into a smartphone-based UI using Lo-Fi paper prototypes.


Using a guerilla testing method, I collected feedback from non-specific users, aiming to identify usability issues and mobile-specific heuristics. Although the volunteers didn't perfectly match the persona profile, their input helped uncover initial design problems.


I conducted six guerilla sessions, and audio-recorded for later review. Observational notes on verbal cues and emotional reactions using the 'Said / Did / Thought / Felt' framework provided valuable insights during the testing process. The rationale behind guerilla testing was its efficient and immediate feedback-gathering approach.


Second iteration


The feedback gathered during the  earlier testing sessions using a Lo-Fi prototype informed the subsequent wireframes designs. The set below shows a happy-path whereby first-time users interact with the system, choosing to fully register and provide further personal details in order to get a tailored experience from the voice-agent surface.

Screen 1.jpg
Screen 3.jpg
Screen 5.jpg
Screen 2.jpg
Screen 4.jpg
Screen 6.jpg


Dialogueflow prototype

To build NoraCare, a multi-modal system based on voice interactions, the engineering team developed a working prototype using an open-source platform with a pre-trained model. They customised a scraper for specific tasks on top of it. Choosing an open-source platform allowed for a quicker start and avoided substantial upfront investment. Dialogueflow by Google was selected as the initial software for prototype development and preliminary tests using Google Home and Google Mini devices.


However, to meet the requirements of dementia-related use cases, a 'Custom Relevance Model' webhook was created to enhance intent fulfillment beyond Dialogueflow's default capabilities. Assisted learning techniques, including overriding tools and manual content ingestion, were implemented to improve the agent's accuracy.


Social Proof

Obtaining further social proof regarding cultural contexts and use cases for voice interfaces in caregiving is crucial. Further user research and collaboration with dementia-oriented organisations will provide insights and uncover further blindspots about caregiver needs and preferences.


Sustainable Business Model


Developing a sustainable business model is essential. I will need to explore different revenue streams, pricing strategies, and partnerships with healthcare providers or insurance companies to establish a viable financial foundation to support development and growth.



Building a complete product/service in this domain requires significant resources. Extensive collaboration with experts in machine learning, UX design, and domain experts is needed to deliver a fully validated, market-ready MVP.

In summary, gathering social proof, establishing a sustainable business model, and allocating necessary resources are critical for the success of a voice-user interface product in the caregiving domain.

Next steps

  1. Conduct further user research with informal caregivers to uncover additional insights and unmet needs.
  2. Partner with dementia-oriented organizations to access necessary databases for training and improving the voice agent.
  3. Refine and train the voice agent based on insights and caregiver feedback.
  4. Conduct iterative testing and continuously improve the voice agent to meet evolving caregiver needs.
  5. Develop a targeted marketing strategy and execute a coordinated launch to create awareness among caregivers


Product concept: Vilmar Pellisson 
Research: Vilmar Pellisson
UX/UI Design: Vilmar Pellisson
Full-stack Development: / Carlos Calvo
Branding: Vilmar Pellisson
Prototype video: Vilmar Pellisson, Karin Haussmann
bottom of page