Avatar with A.I - 1

Hello! How can I help you?

import os

os.system('cls' if os.name == 'nt' else 'clear')

import cv2

import numpy as np

import pygame

import time

import pyttsx3

import threading

import speech_recognition as sr

from PIL import Image, ImageSequence

import io

import base64

import pygetwindow as gw

import pyautogui

import time

import torch

from transformers import GPT2LMHeadModel, GPT2Tokenizer

model_name = "gpt2"

tokenizer = GPT2Tokenizer.from_pretrained(model_name)

tokenizer.pad_token = tokenizer.eos_token  # Χρήση του eos_token ως pad_token

gpt2_model = GPT2LMHeadModel.from_pretrained(model_name)

device = torch.device("cuda" if torch.cuda.is_available() else "cpu")

gpt2_model.to(device)

def generate_text(prompt, max_new_tokens=50):

    inputs = tokenizer(prompt, return_tensors="pt", padding=True, truncation=True, max_length=512)

    input_ids = inputs.input_ids.to(device)

    attention_mask = inputs.attention_mask.to(device)

    output = gpt2_model.generate(

        input_ids,

        attention_mask=attention_mask,

        max_new_tokens=max_new_tokens,

        num_return_sequences=1,

        no_repeat_ngram_size=1,

        do_sample=True,

        top_k=50,

        top_p=0.90,

        temperature=0.8,

        pad_token_id=tokenizer.eos_token_id,

        eos_token_id=tokenizer.eos_token_id

    )

    raw_text = tokenizer.decode(output[0], skip_special_tokens=True)

    lines = raw_text.split("\n")

    filtered_lines = []

    for line in lines:

        if line.strip().startswith("User:") or line.strip().startswith("AI:"):

            continue

        filtered_lines.append(line)

    final_text = " ".join(filtered_lines).strip()

    return final_text

def build_prompt(conversation_history, new_user_input):

    system_message = (

        "You are beautiful.\n\n"

    )

    dialogue = ""

    for i, msg in enumerate(conversation_history):

        role = "User" if i % 2 == 0 else "AI"

        dialogue += f"{role}: {msg}\n"

    dialogue += f"User: {new_user_input}\nAI:"

    return system_message + dialogue

recognizer = sr.Recognizer()

mic = sr.Microphone()

def listen():

    with mic as source:

        print("🎤 Speak now...")

        recognizer.adjust_for_ambient_noise(source, duration=1)

        try:

            audio = recognizer.listen(source, timeout=5)

            print("✅ Audio Captured!")

        except sr.WaitTimeoutError:

            print("⏳ No speech detected. Try again.")

            return None

    try:

        text = recognizer.recognize_google(audio, language="en-US")

        print(f"✅ Recognition: {text}")

        return text

    except sr.UnknownValueError:

        print("🤖 Could not understand.")

    except sr.RequestError:

        print("❌ Connection error.")

    return None

engine = pyttsx3.init()

engine_lock = threading.Lock()

voices = engine.getProperty('voices')

female_voice_id = None

for voice in voices:

    name_lower = voice.name.lower()

    if "female" in name_lower or "zira" in name_lower:

        female_voice_id = voice.id

        break

if female_voice_id:

    engine.setProperty('voice', female_voice_id)

else:

    print("❌ No female-like voice found. Using default.")

def speak(text):

    with engine_lock:

        engine.say(text)

        engine.runAndWait()

def create_avatar_animation(image_path):

    if not os.path.exists(image_path):

        print(f"❌ File {image_path} not found!")

        exit()

    gif = Image.open(image_path)

    frames = [

        pygame.image.fromstring(frame.convert("RGBA").tobytes(), frame.size, "RGBA")

        for frame in ImageSequence.Iterator(gif)

    ]

    return frames

image_path = r"C:\Users\tsifa\Desktop\avatar 111.webp"

avatar_frames = create_avatar_animation(image_path)

frame_index = 0

pygame.init()

pygame.font.init()

window_open = False

running = True

avatar_text = "Hello! How can I help you?"

def bring_window_to_front():

    time.sleep(0.5)  # Δίνουμε λίγο χρόνο στο Pygame να δημιουργήσει το παράθυρο

    win = gw.getWindowsWithTitle("AI Assistant")  # Βρίσκουμε το παράθυρο

    if win:

        win[0].activate()  # Το ενεργοποιούμε

        pyautogui.click(win[0].left + 10, win[0].top + 10)  # Προαιρετικό κλικ

def update_avatar_text(new_text):

    global avatar_text

    avatar_text = new_text

    pygame.event.post(pygame.event.Event(pygame.USEREVENT))

def show_avatar():

    global window_open, running, frame_index, avatar_text

    if window_open:

        return

    window_open = True

    screen = pygame.display.set_mode((500, 500))

    pygame.display.set_caption("AI Assistant")

    font = pygame.font.SysFont("Arial", 24)

    bring_window_to_front()

    while running:

        screen.fill((0, 0, 0))

        text_surface = font.render(avatar_text, True, (155, 155, 155))

        screen.blit(text_surface, (50, 450))

        current_frame = avatar_frames[frame_index]

        screen.blit(pygame.transform.scale(current_frame, (400, 400)), (50, 50))

        pygame.display.update()

        frame_index = (frame_index + 1) % len(avatar_frames)

        time.sleep(0.1)

        for event in pygame.event.get():

            if event.type == pygame.QUIT:

                print("🛑 Exiting program...")

                running = False

                pygame.quit()

                os._exit(0)

def conversation_loop():

    global running

    conversation_context = []

    # Αρχικό μήνυμα

    initial_message = "Hello! How can I help you?"

    print(f"🤖 AI: {initial_message}")

    update_avatar_text(initial_message)

    speak(initial_message)

    while running:

        print("🎤 Waiting for user input...")

        user_input = listen()

        if user_input is None:

            print("🔄 No input detected. Listening again...")

            continue

        if user_input.lower() in ["quit", "exit", "stop"]:

            print("🛑 Exiting...")

            running = False

            pygame.event.post(pygame.event.Event(pygame.QUIT))

            break

        prompt = build_prompt(conversation_context, user_input)

        ai_response = generate_text(prompt)

        print(f"📝 User: {user_input}")

        print(f"🤖 AI: {ai_response}")

        update_avatar_text(ai_response)

        speak(ai_response)

        conversation_context.append(user_input)  

        conversation_context.append(ai_response)

        if len(conversation_context) > 6:

            conversation_context = conversation_context[-6:]

def detect_motion(threshold=5000):

    cap = cv2.VideoCapture(0)

    time.sleep(2)

    ret, frame1 = cap.read()

    ret, frame2 = cap.read()

    print("🔍 Waiting for motion...")

    motion_detected = False

    while True:

        diff = cv2.absdiff(frame1, frame2)

        gray = cv2.cvtColor(diff, cv2.COLOR_BGR2GRAY)

        blur = cv2.GaussianBlur(gray, (5, 5), 0)

        _, thresh = cv2.threshold(blur, 20, 255, cv2.THRESH_BINARY)

        dilated = cv2.dilate(thresh, None, iterations=3)

        contours, _ = cv2.findContours(dilated, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)

        total_area = sum(cv2.contourArea(contour) for contour in contours)

        if total_area > threshold:

            print("🚨 Motion detected! Starting program...")

            motion_detected = True

            break

        frame1 = frame2

        ret, frame2 = cap.read()

        if not ret:

            break

        if cv2.waitKey(10) & 0xFF == ord('q'):

            break

    cap.release()

    cv2.destroyAllWindows()

    return motion_detected

if detect_motion():

    conversation_thread = threading.Thread(target=conversation_loop)

    conversation_thread.start()

    show_avatar()

    conversation_thread.join()

else:

    print("❌ Motion detection failed or interrupted.")