tevr-asr-tool VS MathAnimation

Do some cool open source stuff. Since August 9th, I received about 50 job offers mentioning this repo: https://github.com/DeutscheKI/tevr-asr-tool Most of them were senior engineer or AI researcher, with a few CTO / co-founder offers sprinkled in. I'm not in the market and this was a bit unexpected to me, but those emails sounded like they would pay well. And most was remote for US companies.

Me, because I plan to extend to realtime processing: https://github.com/DeutscheKI/tevr-asr-tool

Also, all the performance critical stuff on my production servers is C++ with JNI or pybind wrappers.

The unique work that makes this speech recognition superior to other tools is in those 284 lines of code: https://github.com/DeutscheKI/tevr-asr-tool/blob/master/tevr...

That's a custom-designed beam search decoder implemented in C++ and based on the research for my TEVR paper. It increases performance by a relative 16% reduction in word error rate.

https://github.com/ambrosiogabe/MathAnimation

If you’ve ever tried to make a mathematical animation (think 3Blue1Brown), it’s a real pain. I was using manim for awhile to make animations for my YT channel, but the whole iteration process felt very slow and repetitive. So I thought I would recreate manim over the weekend, except with a GUI and real-time feedback. It’s been a year and a half and I’m hoping this weekend will be done soon so I can move on and start making videos again.

So far, it does a lot, but it still needs a lot of polish and refinement. The readme gives some gifs and a better idea of the feature set right now.

Win32 definitely has some stinkers. But, video encoding APIs definitely take the cake for me. I've only tried a couple of FFmpeg's APIs and AV1's API, but my God, these are the worst API's I've ever had to deal with.

Just as an example, all I wanted was an API like this[0] for FFmpeg. In order to implement that API (which in my opinion is reasonable), I had to write this monstrosity[1]. It took me a solid week to find an example of how to do this, then another few days of fiddling until I finally just barely got something working. Then I threw in the towel even though the performance was horrible. I tried again a year later and spent another month wrestling with AV1 :/

The amount of leakage going on in these APIs is absolutely insane. I shouldn't have to know the intimate details of how video encoding works to use your library. If I do, then I may as well write my own encoder at that point.

[0]: https://github.com/ambrosiogabe/MathAnimation/blob/18c004bca...

[1]: https://github.com/ambrosiogabe/MathAnimation/blob/18c004bca...

I just so happen to be working on a real-time Gui first replacement for manim :)

It still has a ways to go, but I was able to create one video with it so far and I'm working on all the pain points I ran into while using it. Feel free to check it out if you're interested!

https://github.com/ambrosiogabe/MathAnimation

I'm using C++ for a minecraft clone that I've been tinkering on for the past year[0]. I also plan on using embedded lua for scripting, and I'm using RML UI for game HUDs, ImGui for development tools, and OpenGL for graphics. I use premake for my build system but plan on switching to CMake.

I'm also using it for an animation tool[1]. I've been using 3Blue1Brown's Manim (written in Python) which is amazing, but it lacks real-time editing and proper 3D blending. It also lacks audio synchronization, 3D texture support, and some more complex features that I'd like to add :)

[0]: https://youtu.be/UAUdIQZKV88

[1]: https://github.com/ambrosiogabe/MathAnimation