I would like to have the ability to fill a text field with speech to text. Say, if a user could press a button and say "pencil", then a text box or input field would then read pencil. Is there a way to do this? I have a USB Microphone.
You can use Windows' own speech-to-text features to enter text into fields in a Power App. The user needs to activate the feature, which is available if you search for Windows Speech Recognition on the Start bar.
Hope this helps,
A quick online search brings up several speech-to-text options for Raspberry Pi, but have a look too at the Azure Speech to Text service or the Google Cloud Speech to Text service.