app.py

import gradio as gr
import time
from transformers import pipeline

p = pipeline("automatic-speech-recognition", 
             model="carlosdanielhernandezmena/wav2vec2-large-xlsr-53-faroese-100h")

def transcribe(audio, state="", uploaded_audio=""):
    if uploaded_audio:
        audio = uploaded_audio
    time.sleep(3)
    text = p(audio)["text"]
    state += text + " - "
    return state, state

gr.Interface(
    fn=transcribe, 
    inputs=[
        gr.inputs.Audio(source="microphone", type="filepath"),
        'state',
        gr.inputs.Audio(label="Upload Audio File", type="numpy", source="upload")
    ],
    outputs=[
        "textbox",
        "state"
    ],
    live=True).launch()