Author profile picture

From text messages to video calls, virtual communications have become firmly embedded in most social interactions. In fact, for many people, it is almost inconceivable nowadays to imagine getting through the day without them. And this has never been more true than in the current COVID-19 environment where most of us completely rely on these communications in our day-to-day lives.

However, a lot can be lost during the process of replacing face-to-face interactions with virtual communications. Didimo, a Portugal-based start-up, aims to bring the richness of human communication to every online interaction.

Verónica Orvalho, CEO and founder of Didimo, explained to Innovation Origins what the motivation behind her company is.

“I clearly saw that the ways in which we connect with each other are evolving, but there is one thing that is missing. How can I connect virtually in a more empathetic way? Because I truly believe that sending text messages usually causes lots of misunderstandings. So, it’s about building a new technology that would allow us to communicate on a more human level via a digital device.”

That’s why Didimo has created a software that allows users to have a lifelike digital version of themselves in their virtual communications. This offers them the opportunity to incorporate human attributes that are at the core of human connection and communication: visual appearance and animation, behavior, emotion, and voice.

For example, a Didimo avatar could read your emails and transmit the right emotion that might otherwise be lost in a text.

Didimo Team. ®Didimo

The technology

Didimo already has an application programming interface (API) available for companies to use. The CEO stated that developers use this technology to integrate lifelike digital humans, or Didimos, in their own applications.

But how does it work and how much can the Didimos actually resemble their real-life counterpart?

The available technology uses pictures to recreate the likeness of the users. People can take a selfie or upload several photos, which will allow for more in-depth information, and their Didimo will be created based on that data. The more information provided, the better the depiction the Didimo will be of the user.

Matt Vernon-Clinch, Director of Business Development at Didimo, explained that the technology brings peer to peer interactions back to any number of different mediums.

“We take photographic data of people into the pipeline, and obviously there are numerous ways you can input that. But what comes out of our process is an optimized 3D real-time clean topology mesh with a real-time rig and it’s ready for animation.” He added that the pipeline includes typical mouth shapes and facial expressions that people make when they are talking.

Motion capture update on the way

He added that what comes out from this automated process is what is referred to as a 3D asset, that can be implemented into multiple products or services that the end-user has.

“This is a very scalable product; people may want it for small-scale character generation. But we’re also looking at the very large-scale platform usage where we can be an integrated part of somebody else’s platform.”

Although the API its the company’s core product, there is an app available on GooglePlay as well. On it, users can create their own Didimo and make them speak through text-to-speech technology. A new update will be available soon that will include motion capture. This allows users to easily recreate movement, which puts tech into the hands of anyone something that has so far been unaffordable.

Didimo App. ®Didimo

Getting there

Orvalho has over 15 years of experience in character and spatial animation mainly within the gaming and film industries. It was back then when the technology for Didimo first started to be developed.

Always interested in research and academia and considering how time-consuming and expensive the character creation process was, Orvalho asked herself at some point how she could automate this process.

With this goal in mind, she set up a lab with approximately 20 researchers and engineers. Despite being mainly focused on the gaming industry, the next thing she knew she was being invited to participate in EU projects in the healthcare sector. She has brought her animation technology to studies on embodied cognition that looked into how empathy levels could be improved when dealing with domestic violence victims, help physically disable people to recover damaged areas inside their brains and help autistic children improve their communication skills.

All her years of research left her with a ton of technology for automating 3D characters and drawing expressions on them. However, Orvalho said that it wasn’t until 2010 when this platform came onto the market that she saw her research with entirely different eyes.

“The father of one of the autistic children that were coming to try our test application asked if he could take the platform with him. And I asked why for him and not for his son. And he said that he wanted to use the platform so that he could communicate with his son.”

“Suddenly, for the first time, I felt that we were developing could impact others. Because that was always on my mind, how this technology could actually be used to improve communication.”


One of the main challenges to create this technology was something that Orvalho has been working on for several years: how to avoid the uncanny valley.

The concept of the uncanny valley in robotics was first introduced by Japanese Masahiro Mori, who argued that people would reject humanlike robots, whose appearance resembled yet did not quite mimic perfectly a real human being.

Didimo managed to cross the uncanny valley, which was essential for the success of the technology.

The next challenge is in the application of movement for the Didimos. However, the technology has already been developed and is already available, but has not yet been rolled out on a major scale.

Nonetheless, Orvalho expects that in future people will use Didimo as their interface to connect with other people in a more human and empathetic manner. Additionally, she hopes her company and technology can contribute to the ways children in different cultures are educated and can communicate.