app.py

import gradio as gr
from transformers import pipeline



device = "cuda:0" if torch.cuda.is_available() else "CPU"

def transcribe(audio):
    
    pipe = pipeline(
      "automatic-speech-recognition",
      model="openai/whisper-small",
      chunk_length_s=30,
      device=device,
    )
    
    ds = load_dataset("hf-internal-testing/librispeech_asr_dummy", "clean", split="validation")
    sample = ds[0]["audio"]
    
    prediction = pipe(sample.copy(), batch_size=8)["text"]
    " Mr. Quilter is the apostle of the middle classes, and we are glad to welcome his gospel."

    return prediction

gradio_app = gr.Interface(
    prediction,
    inputs=gr.Image(label="Input", sources=['audio'], type="pil"),
    outputs=[gr.Image(label="Ouput"), gr.Label(label="Result", num_top_classes=2)],
    title="Transcribed",
)


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