app.py

from transformers import pipeline
import gradio as gr

pipe = pipeline(model="Abosteet/XLSR-Wav2Vec2-arabic")

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

iface = gr.Interface(
    fn=transcribe,
    inputs=gr.Audio(type="filepath"),
    outputs="text",
    title="The Arab XLSR Wav2Vec2",
    description="Arabic speech recognition using a fine-tuned Wav2Vec2-XLSR-53 model on subsets from Arabic Speech Corpus and Common Voice 17.0 datasets.",
)

iface.launch(inline=False)