At the beginning of this week I spent a bit of time doing some final preparation for my design presentation. Also towards the start of the week I did some work with David with signal generation on the Pi. Because of the change in how the GPIO pins work on the Pi 5, none of the libraries seemed to have the functionality that we needed (hardware timed and able to create a wave of multiple pulses at once) to achieve the accuracy we want, so we were testing a method using a Linux kernel driver for IR that had mixed reviews on the Pi forums. This didn’t work so we are migrating our design to use a Pico to handle the IR hardware interfacing.
At the end of the week I pivoted and got the Vosk speech to text working on the Pi with our breakout mic.
I am still on schedule if not slightly ahead again. The Vosk took less time to integrate with the Pi than I had anticipated, and I’ll be able to get back to signal generation when the Picos arrive next week so the schedule is looking good.
Next week i’ll be spending some time on the design report document. I’ll also be working (along with David) on the signal generation and decoding code on the Pico as well as the code to accurately query Vosk.