I built a project called Resonant - it's a local speech to text tool using NVIDIA Parakeet and supports a few other languages. Originally started as an experiment to see what could I, and how far could I go building with LLMs to replace my own software stack.
I spent a lot of time on positioning and design. The thing is like 10% model 90% unique, random, obscure features I had to run into walls for days to identify how to solve. Even with AI.
Lots of respect to the leaders in this space. So I ended up open-sourcing the core.
After many requests for dozens of edge features and active engagement into the shape of the product, given where software development is and coding agents are right now, I've decided to open source it to allow people the flexibility to manage and personalize their speech to text pipeline on their own. Resonant is now open source!
An IDE for agents, that includes a general purpose agent inside so developers don't need to start with a framework or from scratch. It just works, out of the box.
I feel that the software development lifecycle for agents is uniquely iterative and requires a different set of tools and environment, so that's how this came to be. First we built the general purpose agent -> now we build the IDE around it.
I spent a lot of time on positioning and design. The thing is like 10% model 90% unique, random, obscure features I had to run into walls for days to identify how to solve. Even with AI.
Lots of respect to the leaders in this space. So I ended up open-sourcing the core.
https://github.com/tohmsc/resonant-community