app.py

import gradio as gr
from transformers import pipeline

trans = pipeline("automatic-speech-recognition", model = "openai/whisper-small")


def audio2text(audio):
  text = trans(audio)["text"]
  return text

gr.Interface(
    fn=audio2text,
    inputs = [gr.Audio(source="microphone", type="filepath")],
    outputs=["textbox"]
).launch()