Jannis Conen

Hate creating videos? Not anymore! Introducing Vidify🚀

Created October 7, 2023

Creating explainer videos takes hours and if I make one mistake I can do it all over again!

I have a study platform for students in Germany and had to make creating videos easy, if I ever wanted to get more users. Videos were the #1 request on the platform. So I came up with an idea that reduced the time effort of creating videos on my platform by 80%! Today we launched it as a service on https://vidify.cloud 🚀

When a professor explains something in a lecture, they use slides and talk over them. I had just seen the awesome library tldraw by Steve Ruiz, an infinite canvas in React with stellar handwriting. So I came up with the idea to create a web-based presentation editor that turns your presentations into videos by adding a synthetic voice and rendering the combination of slides and speech to a video. This allows quick creation and better maintainability because you could just rerender if there was any mistake. 🤯

video_editor.png

I used to create math videos and was furious when I received the feedback that I had a miscalculation at minute 1:02, so I could redo it all over again, and god did I hate recording myself at the time!

How did I build it?

I took tldraw for the frontend and looked at other presentation editors like PowerPoint. The goal was to make it as simple and clean as possible. I added a textarea and the option to change the language of the speaker. When saving a video, it would be processed in Python using FFMPEG. At first I created the voice with Google Cloud tts (bad idea :D)

The feedback

Other tutors loved it! I myself from then on only created videos using it because it was such a time saver and the ability to collaborate with other tutors who would just edit the video and rerender it was awesome!

BUT... our users did not love it. The Google cloud tts was (and mostly still is) awful, especially the German voices.

Super improvements 💪🏻

I had to switch to AWS and improve the FFMPEG commands to fix the voice problem. I also added the option to just record your voice for every slide and have a small audio editor. audio_editor_modal.png

Thank you 😊

I'm an economist turned IT-nerd, and this is my first blog post and first SaaS. Thank you so much for reading this post. And a special thanks to my intern, Basti, who was the reason we extracted the video editor and turned it into something usable for everyone! I hope you have a great experience trying Vidify and leave some feedback

Fun Fact: Steve from tldraw was also the reason I started learning React last year, but that story is for another post