Let me preface this by saying; im still a total noob at bot building. but im learning as i go! That being said; As the title suggests, i'd love to see an option for listening for audio as a way of gathering input for a conversation. More specifically, have users press and hold a button to record something they say, which then has the option to feed into S2T and subsequently be used as a user question for the model to respond to. One of my use-cases deals with people with low literacy and/or low tech skills (e.g. elderly), so I'm trying to limit the amount of textual input needed, and this would be a great way to cover that 'problematic' end not covered by things like a Narrator. Thanks for all your hard work!🐐