app.py

import gradio as gr
from transformers import pipeline,WhisperProcessor, WhisperForConditionalGeneration
import torch
import librosa

checkpoint = "openai/whisper-base"
# checkpoint = "/innev/open-ai/huggingface/openai/whisper-base"
processor = WhisperProcessor.from_pretrained(checkpoint)
model = WhisperForConditionalGeneration.from_pretrained(checkpoint)
text_Interface=gr.Interface.load("models/nlpconnect/vit-gpt2-image-captioning")

def greet():
    return "Hello "

with gr.Blocks() as demo:
    gr.Markdown("Start typing below and then click **Run** to see the output.")
    with gr.Row():
        inp = gr.Image(type='pil')
        out = gr.Textbox()
    gr.Interface(fn=greet, inputs=inp, outputs=out)

demo.launch()
text_Interface.launch()