app.py

from transformers import pipeline
import gradio as gr

pipe = pipeline(model="bhuang/wav2vec2-xls-r-1b-cv9-fr")

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

iface = gr.Interface(
    fn=transcribe, 
    inputs=gr.Audio(source="microphone", type="filepath", label="Record something..."), 
    outputs="text",
    title="Automatic Speech Recognition in French",
    description="Realtime demo for French automatic speech recognition using a fine-tuned wav2vec2 model.",
)

iface.launch()