app.py

import speech_recognition as sr  # type: ignore
import os
import pygame  # type: ignore
import gradio as gr
from gtts import gTTS  # type: ignore

# Dog sound files (Ensure these files exist in the same directory)
dog_sounds = {
    "sit": "dog_sit.mp3",
    "come": "dog_come.mp3",
    "fetch": "dog_fetch.mp3",
    "treat": "dog_treat.mp3",
    "play": "dog_play.mp3",
    "bark": "dog_bark.mp3",
    "angry": "mixkit-angry-and-agitated-dog-growling-53.wav",
    "happy": "mixkit-happy-puppy-barks-741.wav",
}

# Initialize speech recognizer
recognizer = sr.Recognizer()
os.environ["SDL_AUDIODRIVER"] = "dummy"  # Prevents pygame audio errors in headless mode
pygame.mixer.init()

def recognize_speech(audio_file):
    """Recognizes speech from an uploaded audio file."""
    try:
        with sr.AudioFile(audio_file) as source:
            audio = recognizer.record(source)
            command = recognizer.recognize_google(audio)
            return command.lower()
    except sr.UnknownValueError:
        return "Sorry, I could not understand your speech."
    except sr.RequestError:
        return "Sorry, the speech service is unavailable."
    except Exception as e:
        return f"Error: {str(e)}"

def dog_response(command):
    """Plays the corresponding dog sound and generates a speech response."""
    if command:
        for key in dog_sounds:
            if key in command:
                sound_file = dog_sounds[key]
                play_dog_sound(sound_file)
                return sound_file, f"Playing sound for {key}", generate_speech(f"Woof! I heard you say {key}")
        
        # If no specific command is recognized, play bark
        sound_file = dog_sounds["bark"]
        play_dog_sound(sound_file)
        return sound_file, "No specific dog command recognized. Playing default bark sound.", generate_speech("Woof! I didn't recognize that, so I'll just bark!")
    
    return None, "No command to process.", None

def play_dog_sound(sound_file):
    """Plays an audio file using Pygame."""
    if os.path.exists(sound_file):
        pygame.mixer.music.load(sound_file)
        pygame.mixer.music.play()
        while pygame.mixer.music.get_busy():
            continue
    else:
        return f"Error: Sound file '{sound_file}' not found."

def generate_speech(text):
    """Generates a TTS response and returns the file path."""
    speech_file = "dog_response.mp3"
    tts = gTTS(text=text, lang="en")
    tts.save(speech_file)
    return speech_file  # Return audio file for Gradio

def process_command(audio_file):
    """Processes the command from the user."""
    command = recognize_speech(audio_file)
    sound_file, response_text, speech_file = dog_response(command)
    
    # Return the dog sound file in the first output
    return sound_file, response_text, speech_file

# Gradio UI with a purple theme
custom_css = """
body {
    background-color: #6a0dad;
    color: white;
    text-align: center;
}
.gradio-container {
    font-family: 'Arial', sans-serif;
    background-color: #800080;
    padding: 20px;
    border-radius: 15px;
}
button {
    background-color: #ffccff !important;
    color: #6a0dad !important;
    font-weight: bold !important;
}
"""

iface = gr.Interface(
    fn=process_command,
    inputs=gr.Audio(sources=["microphone"], type="filepath"),
    outputs=["audio", "text", "audio"],
    title="🐶 Dog Command Recognition 🐶",
    description="🎤 Speak a command and let the dog respond! 🐕\n\nTry commands like 'happy', 'angry', 'sit', 'come', 'fetch', 'treat', 'play'",
    live=True,
    theme="compact",
    css=custom_css,
)

if __name__ == "__main__":
    iface.launch(share=True)