I originally built SquishyFile as a video compressor.
It was a pretty simple idea: take a video, make it smaller, and let me choose the size I wanted. I shared it with a few friends, they actually used it, and some of them told me it was useful. So I put it on a domain and made it public.
Since then, I've kept adding things whenever I run into another video problem worth solving.
The newer stuff is probably more interesting than the original compressor. There’s an AI video upscaler that can turn a clip into a 2x or 4x version, and a video-to-text tool that can transcribe speech in multiple languages.
I also wanted the AI tools to be usable without turning the interface into something complicated. Pick a video, choose what you want, and let it run.
The upscaler was one of the more interesting things to build. Different videos need different approaches, so it doesn't blindly process every source the same way. The video-to-text tool had its own set of problems around getting audio into the format the speech model expects. Those are the kinds of problems I enjoy working on.
The original tools are still there too: video compression, target-size compression, video-to-MP3, GIF creation, and extracting frames from videos.
I didn't really have a big plan for any of this. It just kept growing whenever I had another idea for a small video tool.
It gets around 500 users a day now, which is pretty cool considering it started as something I was making for myself.
It's still a side project. I'm currently interested in adding more AI and video tools, including things like background removal or replacement and more transcription languages.
For now, I'm just going to keep building the things that seem useful.
Top comments (0)