app.py

import gradio as gr
import torch
from transformers import pipeline

device = "cuda" if torch.cuda.is_available() else "cpu"

stt = pipeline(
    "automatic-speech-recognition",
    model="openai/whisper-small",
    device=0 if device == "cuda" else -1
)

def speech_to_text(audio):
    result = stt(audio)
    return result["text"]

app = gr.Interface(
    fn=speech_to_text,
    inputs=gr.Audio(
        sources=["microphone"],
        type="filepath",
        label="Speak"
    ),
    outputs=gr.Textbox(label="Transcribed Text"),
    title="Multilingual Speech to Text AI",
    description="Supports English, Urdu, and 90+ languages"
)

app.launch()