inference_classifier.py

import pickle
import cv2
import mediapipe as mp
import numpy as np
import pyttsx3
import serial
import time

ser = serial.Serial("COM7", 9600) #Change your port name COM... and your baudrate
#ser.open()
#qser.reset_input_buffer()

model_dict = pickle.load(open('./model.p', 'rb'))
model = model_dict['model']

cap = cv2.VideoCapture(1)

mp_hands = mp.solutions.hands
mp_drawing = mp.solutions.drawing_utils
mp_drawing_styles = mp.solutions.drawing_styles

hands = mp_hands.Hands(static_image_mode=True, min_detection_confidence=0.3)

# initialize Text-to-speech engine
engine = pyttsx3.init()

word = ""

#labels_dict = {0: 'A', 1: 'B', 2: 'L'}
while True:

    data_aux = []
    x_ = []
    y_ = []

    ret, frame = cap.read()

    H, W, _ = frame.shape

    frame_rgb = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)

    results = hands.process(frame_rgb)
    if results.multi_hand_landmarks:
        for hand_landmarks in results.multi_hand_landmarks:
            mp_drawing.draw_landmarks(
                frame,  # image to draw
                hand_landmarks,  # model output
                mp_hands.HAND_CONNECTIONS,  # hand connections
                mp_drawing_styles.get_default_hand_landmarks_style(),
                mp_drawing_styles.get_default_hand_connections_style())

        for hand_landmarks in results.multi_hand_landmarks:
            for i in range(len(hand_landmarks.landmark)):
                x = hand_landmarks.landmark[i].x
                y = hand_landmarks.landmark[i].y

                x_.append(x)
                y_.append(y)

            for i in range(len(hand_landmarks.landmark)):
                x = hand_landmarks.landmark[i].x
                y = hand_landmarks.landmark[i].y
                data_aux.append(x - min(x_))
                data_aux.append(y - min(y_))

            

        x1 = int(min(x_) * W) - 10
        y1 = int(min(y_) * H) - 10

        x2 = int(max(x_) * W) - 10
        y2 = int(max(y_) * H) - 10

        # Ensure data_aux has exactly 84 features
        while len(data_aux) < 84:
            data_aux.append(0)

        prediction = model.predict([np.asarray(data_aux)])

        #changed this
        predicted_character = str(prediction[0])

        cv2.rectangle(frame, (x1, y1), (x2, y2), (0, 0, 0), 4)
        cv2.putText(frame, predicted_character, (x1, y1 - 10), cv2.FONT_HERSHEY_SIMPLEX, 1.3, (0, 0, 0), 3,
                    cv2.LINE_AA) 
        
        #letters
        if cv2.waitKey(1) == ord('w'):
            word += predicted_character
            print(word)
        #to add spaces
        if cv2.waitKey(1) == ord('e'):
            word += " "
            print (word)
        #for words
        if cv2.waitKey(1) == ord('q'):
            word += predicted_character 
            print(word)
            #ser.write(word.encode())
            break   

    cv2.imshow('frame', frame)
    cv2.waitKey(1)

print(word)
cap.release()
cv2.destroyAllWindows()
engine.say(word)
#play the speech
engine.runAndWait()
ser.write(word.encode())
time.sleep(0.05)
ser.close()

#tts = gtts.gTTS(word)
#tts.save("hello.mp3")
#playsound("hello.mp3")
 

engine.say("Bark")