Weeks 1 and 2 were about identifying the dependencies needed for Vosk Offline speech recognition API to run.

I have started writing and testing recipes for the meta-offline-voice-agent layer to install those dependencies on the target image. 

More details are in the following post.

This week, and in the coming ones, I will complete writing the recipes for the dependencies and will also start looking at the audio side of the project, and the models required for the Vosk API to work.

Best Regards

Aman Arora