app.py

import gradio as gr
import google.generativeai as genai

# API key input
import os 
token=os.environ.get("TOKEN") 
genai.configure(api_key=token)

safety_settings = [
  {
    "category": "HARM_CATEGORY_HARASSMENT",
    "threshold": "BLOCK_NONE",
  },
  {
    "category": "HARM_CATEGORY_HATE_SPEECH",
    "threshold": "BLOCK_NONE",
  },
  {
    "category": "HARM_CATEGORY_SEXUALLY_EXPLICIT",
    "threshold": "BLOCK_NONE",
  },
  {
    "category": "HARM_CATEGORY_DANGEROUS_CONTENT",
    "threshold": "BLOCK_NONE",
  },
]

def describe_video(pp,video_file):
    try:        

        print(f"Uploading file...")
        #uploaded_video = genai.upload_file(path=video_file)
        url ="https://youtu.be/zfV-IOn7ouI?si=GWn8WyWLUe178L--"
        uploaded_video = genai.upload_file(path=url)
       
        print(f"Completed upload: {uploaded_video.uri}")

        import time

        while uploaded_video.state.name == "PROCESSING":
            print("Waiting for video to be processed.")
            time.sleep(10)
            uploaded_video = genai.get_file(uploaded_video.name)

        if uploaded_video.state.name == "FAILED":
            raise ValueError(uploaded_video.state.name)
        print(f"Video processing complete: " + uploaded_video.uri)

        prompt = pp

        # Set the model to Gemini 1.5 Pro.
        model = genai.GenerativeModel(model_name="models/gemini-1.5-flash-latest",safety_settings=safety_settings)
        # Make the LLM request.
        print("Making LLM inference request...")
        response = model.generate_content(
            [prompt, uploaded_video], request_options={"timeout": 600}
        )
        print(response.text)

        genai.delete_file(uploaded_video.name)
        print(f"Deleted file {uploaded_video.uri}")

        return response.text
    except Exception as e:
        return f"An error occurred: {e}"


# Create the Gradio interface
iface = gr.Interface(
    fn=describe_video,
    inputs=[gr.Textbox(label="Question:", lines=3) ,gr.Video()],
    outputs=gr.Textbox(),
    title="y Video.",
    description="Pose des questions sur la vidéo et obtient une réponse.",
)

# Launch the interface
iface.launch()