Sonantic’s 2020 in review

Request access

Try our hyper-realistic voices

Zeena Qureshi

CEO & Co-founder

February 11, 2021

What a year 2020 has been! 

We started last year with a plan to launch our beta and finished the year off with: 

  • 100+ monthly active customers 
  • multiple patents in tow
  • an incredible waitlist
  • and — the best team we could have ever asked for :)

Here’s a look at our journey over the last 12 months and what made it special for us (& in case you missed it, here’s Sonantic's Origin Story with our mission and vision to give you a rundown of everything we're all about).

Venturing into the unknown

Like everyone else, we were definitely not ready for a pandemic. We had spent 2019 fundraising and developing our product to unveil at the Game Developers Conference 2020 (GDC). We knew that anyone who’s anyone in the gaming industry was going to be there. We had client and investor meetings lined up, a booth waiting to be decorated with Sonantic banners, and all the energy for a 5 day sprint to sell, sell, sell. Due to the pandemic, we had no other choice but to abandon ship and reshuffle all of our plans. We did a scaled down virtual presentation (which can still be watched today in the GDC Vault).  

Simultaneously, we moved out of our office and had the entire team working remotely by early March, well ahead of the first lockdown in London. This was quite the switch up for us as we had never worked remotely as a team before. To our surprise, the team aced it by being even more productive than before and we started hosting weekly remote team lunches, socially distanced park events and monthly game/quiz evenings which brought us closer together. 

We also ensured that all of our actors were set up with home studios to keep them safe so that they could continue working. Pivoting to this remote style for the actors had a substantial amount of challenges. Firstly because the actors were used to recording in professional studios but were now needing to record in their bedrooms and/or closets (not exactly ideal). Secondly, gathering the right type of data to keep developing emotions was also interesting as we found the walls to be so thin for most, that neighbours close by would get worried when actors recorded for sadness and anger. But by the end, the actors and ML researchers did a fantastic job and we were able to deploy core emotions by the end of the summer.

Let’s get emotional

So onwards we went, with our small, but mighty team of 7, and our collective goal of developing a technology with deep human emotions to allow creatives the ability to bring a story to life — we stayed focused. We trained over 200 models and by May, we had an algorithm that could cry. So we created F(AI)th, the First AI that Can Cry which was featured in TechCrunch, The Times, and more. This was huge for us, we proved to ourselves and the world that not only was synthetic deep emotion possible, but that we could replicate one of the hardest emotions for humans: Sadness with crying (a non-speech pattern).

Scaling up 

We followed this up shortly by launching our closed beta program in May for early customers.

Little did we know that the release of F(AI)th would attract hundreds of companies to our beta, all of which wanted access right away to start creating stories as soon as they could. With less than 10 of us as a team, we knew we would have our work cut out for us. So to ensure the best customer experience and the ability to listen to our users & their feedback, we created a waitlist and thanked everyone for their patience. 

As a seed stage startup, we initially had 7 pilot customers. This was definitely manageable but with demand and interest growing, we knew we had to scale up. By year end, we had  200+ companies in our beta, all focused on the Entertainment industry (games, movies and media). To do this, we had to be practical and keep a growth mindset. Be open to different opportunities, partnerships, and projects. While also being open to change. What worked with managing 7 pilots was not going to work with over 200 of them. 

Make it a moonshot

Every month, we made sure to have 1 day, better than all the rest, reserved just for our teammates. A day of fun, where we could all have the freedom to work on a single project for 8 hours. Where we could all make something that we personally have been thinking about that’s not in our roadmap but in line with our company’s north star. A “moonshot day". Cause why not? Regardless how good or bad the project is, there’s nothing to lose, and only results to be gained. Then we can celebrate as a team for both the epic fails that save us time later and the fantastic wins that push us ahead.

Some of our greatest work of all time has come from moonshots. We believe that in order to make strides, we need to allow the freedom to be creative and really listen to each other. That’s why every first Friday of the month since May 2020, we encourage all of our teammates to explore and test ideas even if it feels impossible. (That’s why it’s called a “moonshot.”)

We’re so proud of this team ritual as it’s brought on so many innovations like controllable models, real-time rendering, new marketing avenues, and revamping the entire actor experience.

Product, product, product

Building products is a very iterative process that requires 1-1, in-depth product feedback calls with core users. We made sure to really focus on this to learn what customers wanted. Thanks to the wonderful support of the beta, over 90% of the features we released were all based on user feedback. Amongst some of them were: the Command-Line Interface (CLI), the App, a custom dictionary, timecode positioning for audio files, bulk editing parameters, SSML, CSV import, the ability to generate multiple renders of the same line, more stock voices, and longer renders. We know that we couldn’t have gotten to where we are today without the invaluable feedback from our beta users. For this, we will always be very grateful.

We will be launching in Q1 and can't wait to announce our new updates and projects (no spoilers!). Most importantly, we want to share that we are ready to scale and open up our technology to the entire world.

Game on!


→ Return to blog

Heading

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Suspendisse varius enim in eros elementum tristique. Duis cursus, mi quis viverra ornare, eros dolor interdum nulla, ut commodo diam libero vitae erat. Aenean faucibus nibh et justo cursus id rutrum lorem imperdiet. Nunc ut sem vitae risus tristique posuere.

  • It prevents you to directly style elements of a RT when this RT is bound to the CMS. I hope this limitation will be removed soon, but in the meantime the workaround is pretty easy to handle. Proceed as follow:
  • CMS. I hope this limitation will be removed soon, but in the meantime the workaroun
  • Vents you to directly style elements of a RT when this RT is bound to the CMS. I hope this limitation will be removed soon, but in the meantime the wor