app.py

import os
import tempfile
import shutil
import google.generativeai as genai
import gradio as gr
import requests
import numpy as np
import subprocess
import matplotlib.pyplot as plt
from matplotlib.animation import FuncAnimation
import PIL.Image
from gradio import processing_utils, utils

# Configure Google Gemini API
genai.configure(api_key=os.getenv("GEMINI_API_KEY"))

# Play.ht API keys
API_KEY = os.getenv('PLAY_API_KEY')
USER_ID = os.getenv('PLAY_USER_ID')

def upload_to_gemini(path, mime_type="image/jpeg"):
    file = genai.upload_file(path, mime_type=mime_type)
    return file

def generate_roast(image_path):
    uploaded_file = upload_to_gemini(image_path)
    generation_config = {
        "temperature": 1,
        "top_p": 0.95,
        "top_k": 40,
        "max_output_tokens": 8192,
        "response_mime_type": "text/plain",
    }
    model = genai.GenerativeModel(
        model_name="gemini-1.5-flash-002",
        generation_config=generation_config,
        system_instruction="You are a professional satirist and fashion expert. Roast the profile picture.",
    )
    chat_session = model.start_chat(history=[{"role": "user", "parts": [uploaded_file]}])
    response = chat_session.send_message("Roast this image!")
    return response.text

def text_to_speech(text):
    url = "https://api.play.ht/api/v2/tts/stream"
    payload = {
        "voice": "s3://voice-cloning-zero-shot/d9ff78ba-d016-47f6-b0ef-dd630f59414e/female-cs/manifest.json",
        "output_format": "mp3",
        "text": text,
    }
    headers = {
        "accept": "audio/mpeg",
        "content-type": "application/json",
        "Authorization": API_KEY,
        "X-User-ID": USER_ID
    }
    response = requests.post(url, json=payload, headers=headers)
    if response.status_code == 200:
        audio_path = "output_audio.mp3"
        with open(audio_path, "wb") as audio_file:
            audio_file.write(response.content)
        return audio_path
    else:
        raise ValueError(f"Error: {response.status_code} - {response.text}")

# Generate waveform and overlay with image
def make_waveform_overlay(audio_path, image_path):
    output_video_path = make_waveform(audio_path, bg_image=image_path, animate=True)
    return output_video_path

# Full Gradio Functionality
def process_image(image):
    roast_text = generate_roast(image)
    audio_path = text_to_speech(roast_text)
    final_video_path = make_waveform_overlay(audio_path, image)
    return roast_text, final_video_path

# Gradio Blocks UI
with gr.Blocks() as demo:
    gr.Markdown("# Image Roast and Waveform Video Generator")
    
    with gr.Row():
        image_input = gr.Image(type="filepath", label="Upload Image")
        output_text = gr.Textbox(label="Roast Text")
        output_video = gr.Video(label="Roast Waveform Video")
    
    submit_button = gr.Button("Generate Roast Video")
    submit_button.click(process_image, inputs=image_input, outputs=[output_text, output_video])

# Launch the app
demo.launch(debug=True)