Update app.py

app.py

@@ -8,42 +8,43 @@ from pydub.playback import play
 
 # Dog sound files (Ensure these files exist)
 dog_sounds = {
-    "sit": r'C:\Users\User\Downloads\dog_sit.mp3',
-    "come": r'C:\Users\User\Downloads\dog_come.mp3',
-    "fetch": r'C:\Users\User\Downloads\dog_fetch.mp3',
-    "treat": r'C:\Users\User\Downloads\dog_treat.mp3',
-    "play": r'C:\Users\User\Downloads\dog_play.mp3',
-    "bark": r'C:\Users\User\Downloads\dog_bark.mp3'
+    "sit": "dog_sit.mp3",
+    "come": "dog_come.mp3",
+    "fetch": "dog_fetch.mp3",
+    "treat": "dog_treat.mp3",
+    "play": "dog_play.mp3",
+    "bark": "dog_bark.mp3"
 }
 
 # Initialize speech recognizer
 recognizer = sr.Recognizer()
+os.environ["SDL_AUDIODRIVER"] = "dummy"  # Prevents pygame audio errors in headless mode
 pygame.mixer.init()
 
-def recognize_speech():
-    """Recognizes speech from the microphone input."""
-    with sr.Microphone() as source:
-        try:
-            audio = recognizer.listen(source, timeout=5)
+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 sr.WaitTimeoutError:
-            return "No speech detected. Please try again."
+    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 based on the recognized command."""
+    """Plays the corresponding dog sound and generates a speech response."""
     if command:
         for key in dog_sounds:
             if key in command:
                 play_dog_sound(dog_sounds[key])
-                return f"Playing sound for {key}"
+                return key, f"Playing sound for {key}", generate_speech(f"Woof! I heard you say {key}")
         play_dog_sound(dog_sounds["bark"])
-        return "No specific dog command recognized. Playing default bark sound."
-    return "No command to process."
+        return "bark", "No specific dog command recognized. Playing default bark sound.", generate_speech("Woof! I didn't recognize that, so I'll just bark!")
+    return "unknown", "No command to process.", None
 
 def play_dog_sound(sound_file):
     """Plays an audio file using Pygame."""
@@ -55,42 +56,34 @@ def play_dog_sound(sound_file):
     else:
         return f"Error: Sound file '{sound_file}' not found."
 
-def make_dog_response(command):
-    """Generates a playful dog response using text-to-speech."""
-    try:
-        tts = gTTS(text=f"Woof! I heard you say {command}", lang='en')
-        response_file = "dog_response.mp3"
-        tts.save(response_file)
-        pygame.mixer.music.load(response_file)
-        pygame.mixer.music.play()
-        while pygame.mixer.music.get_busy():
-            continue
-        os.remove(response_file)
-        return f"Generated response: Woof! I heard you say {command}"
-    except Exception as e:
-        return f"Error generating speech: {e}"
+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():
-    command = recognize_speech()
-    response = dog_response(command)
-    tts_response = make_dog_response(command)
-    return command, response, tts_response
+def process_command(audio_file):
+    command = recognize_speech(audio_file)
+    keyword, response_text, speech_file = dog_response(command)
+    return command, response_text, speech_file
 
+# Gradio UI
 iface = gr.Interface(
     fn=process_command,
-    inputs=[],
-    outputs=["text", "text", "text"],
+    inputs=gr.Audio(sources=["microphone"], type="filepath"),
+    outputs=["text", "text", "audio"],  # Added audio output
     title="🐶 Dog Command Recognition 🐶",
-    description="🎤 Speak a command and let the dog respond! 🐕\n\nTry commands like 'sit', 'come', 'fetch', 'treat', 'play', or anything else!",
+    description="🎤 Speak a command and let the dog respond! 🐕\n\nTry commands like 'sit', 'come', 'fetch', 'treat', 'play'",
     theme="default",
     live=True,
     css="""
-    body { background-color: #f0f8ff; }
-    .output-text { color: #ff4500; font-size: 18px; font-weight: bold; }
-    .interface-title { color: #008080; font-size: 24px; font-weight: bold; }
-    .interface-description { color: #2f4f4f; font-size: 16px; }
+    body { background-color: #f8f9fa; text-align: center; }
+    .output-text { color: #ff4500; font-size: 20px; font-weight: bold; }
+    .interface-title { color: #008080; font-size: 26px; font-weight: bold; }
+    .interface-description { color: #2f4f4f; font-size: 18px; }
     """
 )
 
 if __name__ == "__main__":
-    iface.launch()
+    iface.launch(share=True)