Speech Recognition (OpenAI Whisper or nVidia Nemo) as Input

[BradKML]

It would be great to "chat" with the Obsidian Agent instead of needing to type stuff out in the notebook, for English practice for example with a tutor-esque prompt. Assuming that there is no ASR in the OS, would it be possible to integrate that into the plugin?
If not, can ASR be done by writing notes, and then hack-trigger the agent into reading the written transcript and creating a TTS response?