GameFace: Dr Dimitri Palaz
Head of Machine Learning
Dr. Dimitri Palaz, Head of Machine Learning
Speech Graphics comprises over 50 talented individuals who each drive our company forward through their technological, linguistic, operational, and creative experience.
We want to give a face to the many names who strive to deliver our industry-defining facial animation software.
Dr. Dimitri Palaz, our Head of Machine Learning, is the next Speech Graphics player showing their GameFace.
What does a typical working day look like for you?
It depends on the current projects I’m working on, and what requirements there are. I could be meeting with my team to discuss and brainstorm projects, design and run machine learning experiments, design a new feature for the product, or plan and manage research projects.
What has been the proudest moment of your career so far?
During my Ph.D., I was part of the early wave of researchers in speech to adopt and advocate for the deep learning approach, which is now established as the state-of-the-art. The approach was ground-breaking at the time, as it questioned some aspects of the field that were considered as standard. It took quite some time and effort to completely convince the community of its benefit, but it was worth it! I’m very proud that I was part of that.
More recently, I am proud that my work on our facial animation technology is having a significant impact on the video games industry. One example I have in mind is Call of Duty WW2, released in 2017, which was one of the first AAA titles where our tech was used. It enabled the team at Sledgehammer Games to add more dialogues and NPC interactions, strengthening the game’s campaign mode. Another example is the Tomb Raider series, which I am very fond of. I am quite proud to have contributed to these games and all the others.
What’s the first video game you remember playing? What’s your all-time favourite video game?
My first memory is playing the first Super Mario Bros. on the NES. I also have vivid memories of Ducktale on Gameboy.
Regarding all-time favorites - Starcraft 2 is probably there, I’ve played it extensively since its release. It accompanied me during my university years and provided countless hours of entertainment, including e-sports. I also love the Horizon series, Zero Dawn and Forbidden West. I think it has the best gameplay for its genre and I really love the universe and the engaging story that Guerrilla created, as well as the main character.
What are you currently playing?
I’m playing Sea of Thieves regularly, which is a sandbox adventure game. I am impressed by the genius game design by Rare, which cleverly uses multiplayer interaction to create a gameplay with infinite variations, making the game always interesting and fresh, even after hundreds of hours.
What are you reading/watching/listening to at the moment?
I just finished reading the Lady Astronaut book series by Mary Robinette Kowal. It's one of the best science-fiction books I have read in the past 10 years.
What is your experience of working with Speech Graphics?
I joined Speech Graphics more than 6 years ago, in January 2017, when the company was a small start-up (I was the 7th employee). It was an amazing journey to have the company grow from there to what we are now. It had its ups and down, we had a lot of work to do but it was amazing to see the tech improving and becoming more and more adopted by the industry. It was also quite a journey for me, as I transitioned from academia to industry. I learned how to apply machine learning ideas to real-world products, which proved to be a challenging yet rewarding experience.
What I like a lot at Speech Graphics is that we are not afraid to tackle very difficult problems with limited resources, and always manage to find a good solution. For our facial animation technology, this is possible thanks to our incredible and passionate team of researchers, engineers, linguists, and artists working together to create something unique. It’s an immense pleasure to lead and work with this team, and I’m looking forward to seeing what we are going to achieve next.
What was the highlight of 2023 so far?
As a researcher, one of the highlights of the year is the Interspeech conference, which is the world's largest and most comprehensive conference on speech science and technology. This is an occasion to see the latest breakthroughs and chat with other researchers in the field. I think it’s important that we also contribute to the research community, so every year we present a paper. This year was no exception: we presented our paper titled “Emotion Label Encoding using Word Embeddings for Speech Emotion Recognition”, which was well received by the community. I also received a best paper award for a previous publication, which I am very proud of.
For more information on Dimitri’s 2023, check out his website here.
