app.py

import tkinter as tk
import sounddevice as sd
import numpy as np
import speech_recognition as sr
import subprocess
import os
import threading

def execute_command(command):
    """
    Checks for keywords in the recognized text and executes the corresponding command.
    """
    command = command.lower()
    if "notepad" in command:
        result_label.config(text="Opening Notepad...")
        try:
            subprocess.Popen("notepad")
        except Exception as e:
            result_label.config(text=f"Failed to open Notepad: {e}")
    elif "calculator" in command or "calc" in command:
        result_label.config(text="Opening Calculator...")
        try:
            subprocess.Popen("calc")
        except Exception as e:
            result_label.config(text=f"Failed to open Calculator: {e}")
    elif "shutdown" in command:
        result_label.config(text="Shutting down system...")
        os.system("shutdown /s /t 1")
    elif "restart" in command:
        result_label.config(text="Restarting system...")
        os.system("shutdown /r /t 1")
    else:
        result_label.config(text="Command not recognized.")

def record_and_recognize():
    """
    Records audio for a fixed duration (e.g., 5 seconds) and then uses Google Speech Recognition
    to convert it to text. The recognized text is displayed in the GUI and printed to the console.
    """
    sample_rate = 16000  # You may try 44100 if 16000 doesn't work well.
    duration = 5  # seconds

    status_label.config(text="Recording for 5 seconds... Speak now!")
    print("Recording for 5 seconds...")

    try:
        recording = sd.rec(int(duration * sample_rate), samplerate=sample_rate, channels=1, dtype='int16')
        sd.wait()  # Wait until the recording is finished
    except Exception as e:
        status_label.config(text=f"Recording error: {e}")
        print("Recording error:", e)
        record_button.config(state=tk.NORMAL)
        stop_button.config(state=tk.DISABLED)
        return

    status_label.config(text="Stopped recording. Processing audio...")
    print("Recording complete. Processing audio...")

    # Convert the numpy array to bytes
    audio_data_bytes = recording.tobytes()
    sample_width = 2  # int16 = 2 bytes

    recognizer = sr.Recognizer()
    audio_data = sr.AudioData(audio_data_bytes, sample_rate, sample_width)
    try:
        recognized_text = recognizer.recognize_google(audio_data)
        command_label.config(text="You said: " + recognized_text)
        print("You said:", recognized_text)
        status_label.config(text="Processing command...")
        execute_command(recognized_text)
    except sr.UnknownValueError:
        status_label.config(text="Could not understand the audio.")
        print("Could not understand the audio.")
    except sr.RequestError as e:
        status_label.config(text=f"Request error: {e}")
        print("Request error:", e)
    except Exception as e:
        status_label.config(text=f"Error during recognition: {e}")
        print("Error during recognition:", e)
    finally:
        record_button.config(state=tk.NORMAL)
        stop_button.config(state=tk.DISABLED)

def start_recording():
    """
    Called when the Record button is pressed. Disables the Record button,
    enables the Stop button (though in this fixed-duration version it may not be used),
    and starts the recording process in a new thread.
    """
    record_button.config(state=tk.DISABLED)
    stop_button.config(state=tk.DISABLED)  # Not used in fixed-duration mode
    threading.Thread(target=record_and_recognize, daemon=True).start()

# ---------------------------
# GUI Setup with Tkinter
# ---------------------------
root = tk.Tk()
root.title("Voice Command Recorder (Fixed Duration)")
root.geometry("500x300")

# Label to display the recognized speech
command_label = tk.Label(root, text="Your command will appear here", font=("Helvetica", 12))
command_label.pack(pady=10)

# Label for status messages (Recording status, errors, etc.)
status_label = tk.Label(root, text="Press Record and speak", font=("Helvetica", 10))
status_label.pack(pady=5)

# Label to display the result of any executed system command
result_label = tk.Label(root, text="", font=("Helvetica", 12))
result_label.pack(pady=5)

# Button to start recording
record_button = tk.Button(root, text="Record", font=("Helvetica", 12), command=start_recording)
record_button.pack(pady=10)

# (In this fixed-duration version, the Stop button is not used, but we include it for consistency.)
stop_button = tk.Button(root, text="Stop", font=("Helvetica", 12), state=tk.DISABLED)
stop_button.pack(pady=10)

root.mainloop()