I want to be able to clear the input with gradio at the end of completing the inference without any additional click from the user. Currently with live=True, the inference function is triggered when the recording is stopped.
Here is the relevant part of my code:
audio_input = gr.Audio(source="microphone", type="filepath")
demo = gr.Interface(
fn=inference,
inputs=audio_input,
outputs="text",
live=True
)
demo.launch()
- I tried re-initializing the audio component at the end of my inference function. That didn't change anything
- I looked through the docs related to components and all examples I found for other components and couldn't find anything to do this.