Emoface

On Developing Tools For Training Socio-Emotional Skills

Adela Barbulescu
Founder, Emoface
2
Founders
3
Employees
Emoface
from Grenoble
started January 2020
2
Founders
3
Employees
Discover what tools recommends to grow your business!
Discover what books Adela recommends to grow your business!
Want more updates on Emoface? Check out these stories:

Hello! Who are you and what business did you start?๐Ÿ”—

Hey! My name is Adela Barbulescu and I co-founded Emoface. We are a research spin-off that develops tools for training socio-emotional skills for people that have difficulties in social interactions, such as people on the autistic spectrum.

Our flagship product is Emoface Play&Learn emotions, an application that proposes an adapted learning platform in which autistic kids train by playing and interacting with expressive 3D avatars. The application addresses health and education professionals that accompany the kids, such as psychologists and speech therapists, and also family members. This is our first application and we made a โ€œsoftโ€ launch in April 2021, after having been used by more than 600 professionals and parents.

Our next product will represent a visual content creation tool for professionals that uses AI to generate emotional 3D animations. Our first products address autism because it presents a well-identified and urgent need: people on the autistic spectrum have difficulty expressing, recognizing, and managing emotions, which creates serious problems in terms of social inclusion.

To address these difficulties, they must train their social skills with the help of specialized professionals. But professionals have very limited and non-customizable tools, which doesnโ€™t allow for effective training that can be generalized in real situations. Our solution allows the generation of an infinite number of emotional 3D animations, thus allowing specialists to easily illustrate emotions and social situations.

on-developing-tools-for-training-socio-emotional-skills

What's your backstory and how did you come up with the idea?๐Ÿ”—

I am a Computer Science researcher and in 2013 I was doing my Ph.D. in the field of Affective Computing when I first had the idea of creating Emoface. I was studying how people use gestures and speech to express emotions, with the objective of using AI to create emotional avatars that are capable of expressing the analyzed emotions. During this time I found a study conducted by psychology researcher Simon Baron-Cohen, who used videos with people expressing emotions to train emotion recognition for adults with autism.

One thing I cannot stress enough is the importance of communication and finding opportunities to present my company. Word of mouth is what helped me meet my cofounder and our clinical advisor.

This is how I found out about the social interaction deficits in autism. I also realized that the same method of learning could be more effective if it used easily controllable visual content that could be adapted to the users' needs. To learn and generalize a new skill, autistic people require a lot of visual content and developing a tool that generates expressive 3D animations seemed like something useful for health professionals.

I have to add that at this point I had no entrepreneurial and not even industry experience. I had always been passionate about computer vision, games, the use of technology for social impact, especially in the fields of health and education, and about understanding how the brain works. Combined with a shy personality and a history of difficulties in social situations, I clearly empathized with people who faced deficits in social interactions and was very interested in developing new methods for training and evaluating social skills.

I was, therefore, determined to create a solution by applying the tech developed during my studies. After finishing my Ph.D. and postdoc, the project was accepted by an incubator that transfers tech developed in research labs, who accompanied me through the first market studies and helped me train entrepreneurship skills. In 2017 I met my co-founder, Mayra, a UX/UI designer specialized in apps for autism, whose little sister is on the autistic spectrum. Together we created a few prototypes that we could showcase in a few conferences and the response was clear: there is an urgent need for socio-emotional training tools.

Take us through the process of designing, prototyping, and manufacturing your first product.๐Ÿ”—

The process towards releasing our first product was a long one since a lot of research and ergonomy studies were necessary. To develop the application, we collaborated with several researchers and clinical specialists, conducted observation sessions in over 100 structures, and tested with 600 beta users.

Our design process is particularly tedious because the end-user has specific needs: we talk about children with sensory-perceptive sensibilities who learn better using visual stimuli and who shouldnโ€™t be over-stimulated. All details are important, from the expressivity and morphology of the avatars to the color scheme and audio-visual depiction of concepts. One risk presenting a user-specific design to the caretakers, because they have a harder time understanding the interface for the children.

Then there is the question of the nature of our work: evaluating socio-emotional skills is very difficult and the standardized tests used by health professionals are not always adapted for the specific needs of a person on the autistic spectrum. Testing and evaluation are particularly difficult for non-verbal users, i.e. diagnosed people who cannot communicate verbally.

Great help for us was working with autistic adults who helped us make UX UI decisions, by pointing out clear preferences for audio-visual content used in the app. Their testimonials were very useful to better understand the needs and frustrations of both children and adults on the spectrum.

While the app was designed on principles based on numerous studies and we have received very positive feedback from our users, a longitudinal test is required to assess the long-time effects of the application usage. We are currently preparing a clinical study with a cohort of at least 50 children to study the usability and effectiveness for sustained use of the app for more than 2 months.

Furthermore, important strategic decisions still need to be taken, for instance: whether we will present our software as an educational digital tool or a medical device, where intense work will be needed to obtain the required certifications and further validate a business model? These are very important topics for the following period.

on-developing-tools-for-training-socio-emotional-skills
Early days: working on prototyping, first conferences, and promotion materials (all of which are outdated now)

Describe the process of launching the business.๐Ÿ”—

Since our project has a deeptech aspect, our strategy has been to build a strong community while validating its needs and building prototypes. Our discourse is in continuous improvement and we have recently launched our new website, which is more in line with the identity and image we want to diffuse.

Emoface is based on principles such as neurodiversity, tech for good and design for good, learn by playing, participatory design. Our first website was more representative for the medical or research environment, but now we want to highlight the storytelling and ludic side of our solution.

on-developing-tools-for-training-socio-emotional-skills

In terms of finance, we were first accompanied by an incubator and, since the creation of the company last year, we have won several innovation contests that help us obtain R&D subventions. We are in the process of obtaining a bank loan to sustain regular expenses and a one-year budget for marketing and commercial.

Next year we are preparing fundraising to address: recruitment for R&D (of which 250Kโ‚ฌ are already secured), commercial development, and business development for international scaling, marketing, and communication expenses.

Since launch, what has worked to attract and retain customers?๐Ÿ”—

An important step before launching is building a solid community of users and partners. If you are just launching a business, it is essential that the founders are able to pick up the phone and interview potential clients. As I started, this was a particularly difficult step: first of all, as a creator of the product with a technical background, having a commercial/marketing stance felt very weird; added to this, French was my third language and the people interviewed came from a professional background with a specific lexical field.

What helped me kickstart this step was working with a growth hacker who designed a schedule with a discourse we wrote together: I had 3 weeks to interview 10 associations, 10 schools and 10 liberal practitioners. These first contacts revealed essential information and some of them became longtime partners.

Looking back, I can identify several mistakes, among which the initial tendency to do everything on my own and the obsession with details, without having cleared a strategy.

Another step that helped gather qualitative information was publishing on our website an open questionnaire for different segments (professionals, parents, autistic adults, association members).

The next step was creating partnerships with decision-makers and expanding our network by participating in important events in the field. I particularly recall an AI forum during which I gave an interview (unbeknownst to me) to a reporter representing a national press agency. Her article was later re-published in national newspapers which resulted in 3000 new visitors to our website.

Today we are still in the process of launching our first product but we have been through the process of launching a beta test last year. We did it as soon as the first confinement in France was installed and we believe this had a big impact especially on the usage of the app from home.

The growth was steady and we havenโ€™t yet relied on paid apps nor spent much effort on SEO. 30% of the users come from Google search and 30% from recommendations. The rest is an aggregation of social network (20%), press, newsletter announcements from our partners, conferences and workshops.

Last year was very rich in terms of partners and accelerators we worked with, which also gave us access to their networks and helped us carry a PR campaign in Romania. We are now in the process of defining our digital and commercial strategy in order to attain the objectives for this year: reaching at least 2000 new users, testing the business model (mobile store subscription plan) and conducting a clinical evaluation.

How are you doing today and what does the future look like?๐Ÿ”—

We have many projects in progress but the most important are: in the short term, we are dealing with the official launch of the application in French-speaking countries and in Romania.

In the medium term, we will start a new R&D program for the design and clinical evaluation of new products: tools for training more complex emotions and new social skills, by using machine learning algorithms to generate expressive 3D animations and automatic emotion recognition for the user. In the long run, we want to develop solutions for socio-emotional learning for all children and at an international level.

At the moment we are working on creating the digital strategy for the launch but also looking at the overall strategy for the company: will we continue relying on marketing as a digital tool for learning or work towards obtaining certifications as a medical device? This decision is the cornerstone that will shape all the next actions and especially the financial needs and commercial strategy.

Through starting the business, have you learned anything particularly helpful or advantageous?๐Ÿ”—

Starting this business was a life changer for me: it took me from the research world where I would spend most of my time behind a computer, working on a very specific topic, to running a company and holding multiple roles, most of which I had to learn on my own, with no previous idea of what they imply concretely: business developer, finance, project manager and architect, recruiter, marketing and communication, etc.

Looking back, I can identify several mistakes, among which the initial tendency to do everything on my own and the obsession with details, without having cleared a strategy. Luckily, I was accompanied by several incubators and experts that could advise me in decisive moments. Asking for help and advice is a huge asset for those who hold this skill. I also understood the importance of having a mentor. At this point we have several advisors or board members for essential aspects of our business, from the technical and health to finance and strategy.

One thing I cannot stress enough is the importance of communication and finding opportunities to present my company. Word of mouth is what helped me meet my cofounder and our clinical advisor. We also met our scientific partners who will help conduct a clinical evaluation during a conference. Even so, I seldom have trouble keeping in touch with existing partners, either because of lack of time or for waiting too long to send meaningful news. This is work that requires a good organization but it is always very fruitful; it may mean the difference between keeping a trustful relationship and losing a partner.

The same goes for customer support. Another thing Iโ€™m trying to improve is taking too much time to start tasks I donโ€™t like or donโ€™t master very well. The solution Iโ€™m trying to apply is to block a day for that particular task.

What platform/tools do you use for your business?๐Ÿ”—

In terms of production, we decided to use Unity 3D, a cross-platform game engine that allows the use of 3D environment to create games and also mobile deployment. The rest of the pipeline for delivery is covered by Github and Azure. We managed to successfully deliver our initial release after starting agile development using Jira. All things creative are done with Adobe and XD is particularly useful for creating the UI UX flows for the app or website. Daily communication happens either on Slack or Google meet, especially since we work remote. I recently integrated it with Calendly, which made scheduling meetings much easier.

We use Google Workspace and Drive for all documentation and presentations as this works great as collaborative work tools. For finance, roadmap and CRM I still use Excel because I can keep things simple and adaptable to fast changes. We are present on all social media and we plan to introduce many new tools for digital marketing. The basic KPIs are measured with Google Analytics, Playfab, AppStore connect and Google Play Console.

What have been the most influential books, podcasts, or other resources?๐Ÿ”—

Some books may be specific for our domain, but they are an interesting read for anyone. First resources off the top of my mind:

  • For understanding emotions and neurodiversity: Peter Vermeulen, Autism and emotions
  • For understanding the journey and the brain of an extraordinary autistic person: Temple Grandin, Thinking in pictures
  • For understanding how we think through social psychology experiments, Daniel Kahneman: Thinking fast and slow
  • For training negotiation and entrepreneurship skills: Masterclass

Advice for other entrepreneurs who want to get started or are just starting?๐Ÿ”—

With the risk of repeating known ideas: to launch a business, you should be passionate about the proposed idea; it is what you will live 100% of the time, and if you do not strongly believe in what you are doing, it does not make sense to continue. Related to this idea, I do not think that financial gain should be the main motivation of an entrepreneur, but rather a good indicator of the "health" of a business.

It is important for an entrepreneur to surround himself with people as different as possible from whom to learn; to find a mentor, to create a solid network, to always ask "why?" and test their hypotheses. To be successful, an idea or a product with a brilliant technology needs a fit with the identified market.

Are you looking to hire for certain positions right now?๐Ÿ”—

We are currently looking for a marketing manager, a Unity developer and an AI developer, for full time positions. We also look for possible partners in different countries. That wants to meet people who are passionate about the same subjects and who believe in our project.

Where can we go to learn more?๐Ÿ”—

If you have any questions or comments, drop a comment below!