app.py

import os
os.environ["TRANSFORMERS_NO_TF"] = "1"

from transformers import pipeline
import gradio as gr

# Explicitly define the task so HF knows it's ASR
pipe = pipeline(
    task="automatic-speech-recognition",
    model="Devion333/wav2vec2-xls-r-300m-dv"
)

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

demo = gr.Interface(
    fn=transcribe,
    inputs=gr.Audio(sources=["microphone", "upload"], type="filepath"),
    outputs="text",
    title="Wav2Vec2 ASR Demo",
    description="Realtime demo for English speech recognition using Devion333/wav2vec2-xls-r-300m-dv."
)

if __name__ == "__main__":
    demo.launch()