app.py

import gradio as gr
import os
from deep_translator import GoogleTranslator
from PIL import Image
import requests
import io
import time
from groq import Groq
import torch

os.environ['hugging']
H_key = os.getenv('hugging')
API_URL = "https://api-inference.huggingface.co/models/Artples/LAI-ImageGeneration-vSDXL-2"
headers = {"Authorization": f"Bearer {H_key}"}

os.environ['groq']
api_key = os.getenv('groq')
client = Groq(api_key=api_key)


def query_image_generation(payload, max_retries=5):
    for attempt in range(max_retries):
        response = requests.post(API_URL, headers=headers, json=payload)

        if response.status_code == 503:
            print(f"Model is still loading, retrying... Attempt {attempt + 1}/{max_retries}")
            estimated_time = min(response.json().get("estimated_time", 60), 60)
            time.sleep(estimated_time)
            continue

        if response.status_code != 200:
            print(f"Error: Received status code {response.status_code}")
            print(f"Response: {response.text}")
            return None

        return response.content

    print(f"Failed to generate image after {max_retries} attempts.")
    return None

def generate_image(prompt):
    image_bytes = query_image_generation({"inputs": prompt})

    if image_bytes is None:
        return None

    try:
        image = Image.open(io.BytesIO(image_bytes))  # Opening the image from bytes
        return image
    except Exception as e:
        print(f"Error: {e}")
        return None

def process_audio_or_text(input_text, audio_path, generate_image_flag):
    tamil_text, translation, image = None, None, None

    if audio_path:  # Prefer audio input
        try:
            with open(audio_path, "rb") as file:
                transcription = client.audio.transcriptions.create(
                    file=(os.path.basename(audio_path), file.read()),
                    model="whisper-large-v3",
                    language="ta",
                    response_format="verbose_json",
                )
            tamil_text = transcription.text
        except Exception as e:
            return f"An error occurred during transcription: {str(e)}", None, None

        try:
            translator = GoogleTranslator(source='ta', target='en')
            translation = translator.translate(tamil_text)
        except Exception as e:
            return tamil_text, f"An error occurred during translation: {str(e)}", None

    elif input_text:  # No audio input, so use text input
        try:
            translator = GoogleTranslator(source='ta', target='en')
            translation = translator.translate(input_text)
        except Exception as e:
            return tamil_text, f"An error occurred during translation: {str(e)}", None
        # translation = input_text


    # Generate chatbot response
    try:
        chat_completion = client.chat.completions.create(
                messages=[{"role": "user", "content": translation}],
                model="llama-3.2-3b-preview"
            )
        chatbot_response = chat_completion.choices[0].message.content
    except Exception as e:
        return None, f"An error occurred during chatbot interaction: {str(e)}", None

    if generate_image_flag:  # Generate image if the checkbox is checked
        image = generate_image(translation)

    return translation, chatbot_response, image  # Return both chatbot response and image (if generated)

# Custom CSS for improved styling and centered title
css = """
    .gradio-container {
        font-family: 'Georgia', serif;
        background-color: #f5f5f5;
        padding: 20px;
        color: #000000;
    }
    .gr-row {
        box-shadow: 0px 4px 12px rgba(0, 0, 0, 0.1);
        background-color: #ffffff;
        border-radius: 10px;
        padding: 20px;
        margin: 10px 0;
    }
    .gr-button {
        background-color: #8b4513;
        color: white;
        font-size: 16px;
        border-radius: 5px;
    }
    .gr-button:hover {
        background-color: #6a3511;
    }
    .gr-checkbox-label {
        font-weight: bold;
    }
    .gr-image {
        border-radius: 10px;
        box-shadow: 0px 4px 12px rgba(0, 0, 0, 0.1);
    }
    #main-title {
        text-align: center;
        font-size: 28px;
        font-weight: bold;
        color: #8b4513;
    }
"""

with gr.Blocks(css=css) as iface:
    gr.Markdown("<h1 id='main-title'>🖼️ AI Chatbot and Image Generation App</h1>")

    with gr.Row():
        with gr.Column(scale=1):  # Left side (Inputs and Buttons)
            user_input = gr.Textbox(label="Enter Tamil or English text", placeholder="Type your message here...")
            audio_input = gr.Audio(type="filepath", label="Or upload audio (for Image Generation)")
            image_generation_checkbox = gr.Checkbox(label="Generate Image", value=True)

            # Buttons
            submit_btn = gr.Button("Submit")
            clear_btn = gr.Button("Clear")

        with gr.Column(scale=1):  # Right side (Outputs)
            text_output_1 = gr.Textbox(label="English Transcription", interactive=False)
            text_output_2 = gr.Textbox(label="Chatbot Response", interactive=False)
            image_output = gr.Image(label="Generated Image")

    # Connect the buttons to the functions
    submit_btn.click(fn=process_audio_or_text,
                     inputs=[user_input, audio_input, image_generation_checkbox],
                     outputs=[text_output_1, text_output_2, image_output])

    clear_btn.click(lambda: ("", None, False, "", "", None),
                    inputs=[],
                    outputs=[user_input, audio_input, image_generation_checkbox, text_output_1, text_output_2, image_output])

iface.launch()