app.py

import spaces

import gradio as gr
from transformers import pipeline


MODEL_NAME = "bangla-speech-processing/BanglaASR"

print("Loading model...")


pipe = pipeline(
    task="automatic-speech-recognition",
    model=MODEL_NAME,
)


print("Model loaded.")


@spaces.GPU
def transcribe(audio):
    if audio is None:
        return ""

    result = pipe(audio)

    return result["text"]


demo = gr.Interface(
    fn=transcribe,
    inputs=gr.Audio(type="filepath"),
    outputs=gr.Textbox(),
    title="Bangla Speech Recognition",
    description="Bangla ASR using Hugging Face ZeroGPU",
)


demo.launch()