A text based low power walkie talkie SW based on MeshCore. You push PPT talk and it sends the data as a MeshCore text message, and when other people message you it will speak it out. Everything contained offline and no clouds apart from the sky.
After getting rejected my ship I had uploaded my video to youtube instead of slack.
Buttoning up the code. But it’s still pretty messy.
And managed to implement MeshCore for it! 
BC I have copy pasted a lot of function blocks the chode is glued together with a duck tape 
So after getting the ptt -> STT -> TTS working I have though on how to implement the MC functionality. SO I’ve began with getting familiar with the meshcore library by running the included sample sketch which should send a message to the first contact. I ran that and didn’t see anything, after that I made sure that my other device was on top and it worked!
After that I’ve let ChatGPT to do implement a sample code that will let me communicate via the terminal. And it no workie, fixed it and it works great!
Then implemented to the main codebase with copilot. The prompt was the intended function along with a few more things like automatically playing tts file.
Allright so this log is about research and MVP (minimal viable product)
So I stared playing with onyx-om python library and there are also some examples for python. I’ve played with them a bit and it seems that whisper is the king for STT and as for TTS I’m going for piper.
I have modified the STT code from simpler and less acurate model for whisper which is way better and I don’t mind the lag.
And after that I’ve played with different things like other models and TTS.
After that I managed to get implemented this workflow. You press t it starts recording, after release it transcribes it, and with tts piper it converts back to speach. This is meant to prove me the thesis.
So the main core is mostly done, although the code is very messy.
Alright I’ve started researching available tools and started defining the code.
I won’t bore you with all the decisions, but because of modularity I’ve ended up with a wrapper library called Sherpa NX, which seems to do STT and TTS.
For the first prototype I will be using STT whisper tiny and for the TTS I’ll do piper
