Pushing the limits of video encoding with WebCodecs
JSConf India had their very first in-person conference this year(2023) in Bangalore, India. I’m greatly humbled by the opportunity I got to speak about some of the work I have been doing around video encoding on the web for my product – Slant it. It is organized by the same team behind React India and as expected, they brought the same level of attention and care to maintain the quality of the conference.
I want to say a huge thank you to the entire team behind JSConf India. Having spoken to Sahil(one of the main heads of the conference) and understanding the effort it takes to organize a conference like this, it’s truly a commendable job that the team and volunteers have done 💛
About the talk
In this talk, I explored some of the unexplored and niche parts of the web related to Video Encoding on browsers. Video Encoding is a process of converting individual image frames of a video to a final video file (MP4, WEBM, AVI, etc.) that you can play using a video player. While it might seem simple at first with all the tools and libraries that exist out there, it’s actually a deeeeep rabbit hole that never ends! I went from calling simple functions that simply record a canvas element as a video file to understanding the various internal modules of FFmpeg!
My talk also discussed in-depth about a new Web API – WebCodecs which is designed to provide low-level access to video and audio data for encoding/decoding purposes with support for hardware acceleration. This API has very few resources online because of its niche use case and I hope my talk helps in filling this gap. But, WebCodecs is not the only solution to the Video Encoding problem on the web.
Here’s what people had to say about my talk 💛 💛