Added dropdown arrow

Added fake temp sample mid for Audio component

Added pip package with new Gr.Audio

Now using gr.Audio

Add "Output to speakers" option: render MIDI to audio via FluidSynth

Added progress bar and now using the submit not predict

Added Messaged Out

Delete src/backend/gradio_midibridge/templates

Updated packages

Revert the dockerfile

Removed clock messages

Cleaned up code

Clean up code

Fixed issue with each session after processing. Cleaned up CSS

Fixed delay of midi messages

Added demo and build for app