app.py

import gradio as gr
from transformers import pipeline

modelo = pipeline("automatic-speech-recognition", model="facebook/wav2vec2-large-960h-lv60-self")

def transcribe(audio):
  text = modelo(audio)["text"]
  return text

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