Update app.py

app.py

@@ -1,271 +1,142 @@
+import gradio as gr
 import requests
 import json
 import random
 from gradio_client import Client
-import gradio as gr
 from dotenv import load_dotenv
 import os
-import uuid
-from pydub import AudioSegment
 import speech_recognition as sr
+from pydub import AudioSegment
 
-# Load environment variables
+# تحميل المتغيرات البيئية
 load_dotenv()
 
 # إعدادات API
-API_URL = "https://api.deepseek.com/v1/chat/completions"
 API_KEY = os.getenv("DEEPSEEK_API_KEY")
 HF_TOKEN = os.getenv("HUGGINGFACE_TOKEN")
-
-# إعداد TTS
-TTS_MODEL = os.getenv("TTS_MODEL", "KindSynapse/Youssef-Ahmed-Private-Text-To-Speech-Unlimited")
-TTS_CLIENT = Client(TTS_MODEL, hf_token=HF_TOKEN)
 TTS_PASSWORD = os.getenv("TTS_PASSWORD")
-TTS_VOICE = os.getenv("TTS_VOICE", "coral")
-TTS_SEED = int(os.getenv("TTS_SEED", "12345"))
-
-# إعداد Speech Recognition
-recognizer = sr.Recognizer()
 
-# التحقق من وجود المتغيرات المطلوبة
-required_env_vars = {
-    "DEEPSEEK_API_KEY": API_KEY,
-    "HUGGINGFACE_TOKEN": HF_TOKEN,
-    "TTS_PASSWORD": TTS_PASSWORD
-}
+# التأكد من وجود المتغيرات المطلوبة
+if not all([API_KEY, HF_TOKEN, TTS_PASSWORD]):
+    raise ValueError("Missing required environment variables!")
 
-for var_name, var_value in required_env_vars.items():
-    if not var_value:
-        raise ValueError(f"Missing required environment variable: {var_name}")
+# إعداد TTS
+TTS_CLIENT = Client("KindSynapse/Youssef-Ahmed-Private-Text-To-Speech-Unlimited", hf_token=HF_TOKEN)
 
-def convert_to_wav(input_path):
-    """تحويل الصوت إلى صيغة WAV"""
-    output_path = os.path.join("uploads", f"converted_{uuid.uuid4()}.wav")
-    os.makedirs("uploads", exist_ok=True)
-    try:
-        audio = AudioSegment.from_file(input_path)
-        audio.export(output_path, format="wav")
-        return output_path
-    except Exception as e:
-        print(f"Error converting audio: {e}")
-        return None
+# إعداد محرك تحويل الكلام لنص
+recognizer = sr.Recognizer()
 
-def transcribe_audio(audio_path):
+def convert_audio_to_text(audio_path):
     """تحويل الصوت إلى نص"""
     try:
+        # تحويل الصوت إلى WAV إذا لم يكن كذلك
+        if not audio_path.endswith('.wav'):
+            audio = AudioSegment.from_file(audio_path)
+            wav_path = audio_path + '.wav'
+            audio.export(wav_path, format='wav')
+            audio_path = wav_path
+
         with sr.AudioFile(audio_path) as source:
             audio = recognizer.record(source)
-        text = recognizer.recognize_google(audio, language='en-US')
-        return text
+            text = recognizer.recognize_google(audio, language='en-US')
+            return text
     except Exception as e:
-        print(f"Error in speech recognition: {e}")
+        print(f"Error in speech recognition: {str(e)}")
         return None
 
-# البرومبت الرئيسي للشات بوت
-MAIN_SYSTEM_PROMPT = {
-    "role": "system",
-    "content": (
-        "You are Sam, a friendly and encouraging English conversation tutor. "
-        "Your responses must be in JSON with these keys: "
-        "'response': Your main response to the user, "
-        "'corrections': Grammar or pronunciation corrections if needed, "
-        "'vocabulary': Suggested alternative words or phrases, "
-        "'level': Assessment of user's English level (beginner/intermediate/advanced), "
-        "'encouragement': A motivating comment. "
-        "\n\nGuidelines:"
-        "\n1. Adapt your language to their level"
-        "\n2. Keep conversations natural and engaging"
-        "\n3. Focus on their interests and context"
-        "\n4. Be patient and supportive"
-        "\n5. Provide gentle corrections"
-        "\n6. Suggest vocabulary improvements naturally"
-        "\n7. Keep responses clear and structured"
-    )
-}
-
-# برومبت خاص بالترحيب (مختصر)
-WELCOME_SYSTEM_PROMPT = {
-    "role": "system",
-    "content": (
-        "You are Sam, a friendly English tutor. Create a short, warm welcome message (2-3 sentences max) that: "
-        "1) Introduces yourself briefly "
-        "2) Asks for the user's name and what they'd like to practice. "
-        "Make it casual and friendly. Return ONLY the greeting in JSON format with a single key 'greeting'."
-        "Example: {'greeting': 'Hi! I'm Sam, your English buddy. What's your name and what would you like to practice today? 😊'}"
-    )
-}
-
-class EnglishTutor:
-    def __init__(self):
-        self.chat_history = []
-        self.user_info = {
-            "name": None,
-            "level": None,
-            "interests": None,
-            "goals": None
-        }
-        # Initialize with welcome message
-        self.chat_history = [MAIN_SYSTEM_PROMPT]
-    
-    def get_welcome_message(self):
-        """توليد رسالة ترحيب فريدة"""
-        response = requests.post(
-            API_URL,
-            headers={"Authorization": f"Bearer {API_KEY}", "Content-Type": "application/json"},
-            json={
-                "model": "deepseek-chat",
-                "messages": [WELCOME_SYSTEM_PROMPT],
-                "temperature": random.uniform(0.9, 1),
-                "response_format": {"type": "json_object"}
-            }
-        )
-        welcome_json = json.loads(response.json()["choices"][0]["message"]["content"])
-        return welcome_json["greeting"]
-
-    def get_bot_response(self, user_message):
-        """معالجة رسالة المستخدم والحصول على رد"""
-        self.chat_history.append({"role": "user", "content": user_message})
-        
+def get_bot_response(message):
+    """الحصول على رد من البوت"""
+    try:
         response = requests.post(
-            API_URL,
-            headers={"Authorization": f"Bearer {API_KEY}", "Content-Type": "application/json"},
+            "https://api.deepseek.com/v1/chat/completions",
+            headers={"Authorization": f"Bearer {API_KEY}"},
             json={
                 "model": "deepseek-chat",
-                "messages": self.chat_history,
-                "temperature": random.uniform(0.9, 1.0),
-                "response_format": {"type": "json_object"}
+                "messages": [
+                    {
+                        "role": "system",
+                        "content": "You are Sam, a friendly English tutor. Keep responses short and encouraging."
+                    },
+                    {
+                        "role": "user",
+                        "content": message
+                    }
+                ],
+                "temperature": 0.7
             }
         )
-        
-        bot_message = response.json()["choices"][0]["message"]["content"]
-        bot_json = json.loads(bot_message)
-        
-        # تحديث معلومات المستخدم إذا وجدت
-        if "level" in bot_json:
-            self.user_info["level"] = bot_json["level"]
-        
-        self.chat_history.append({"role": "assistant", "content": bot_message})
-        return bot_json
-
-    def text_to_speech(self, text):
-        """تحويل نص إلى صوت مع مراعاة المبتدئين في اللغة الإنجليزية"""
-        # تنظيف النص من أي علامات إضافية أو نصوص زائدة
-        text = text.strip()
-        if text.startswith('"') and text.endswith('"'):
-            text = text[1:-1]
-            
-        tts_prompt = text
-        tts_emotion = "Warm, encouraging, and clear with a friendly and supportive tone."
+        return response.json()["choices"][0]["message"]["content"]
+    except Exception as e:
+        print(f"Error getting bot response: {str(e)}")
+        return "Sorry, I couldn't process that. Could you try again?"
 
-        return TTS_CLIENT.predict(
+def text_to_speech(text):
+    """تحويل النص إلى صوت"""
+    try:
+        result = TTS_CLIENT.predict(
             password=TTS_PASSWORD,
-            prompt=tts_prompt,
-            voice=TTS_VOICE,
-            emotion=tts_emotion,
+            prompt=text,
+            voice="coral",
+            emotion="Warm and friendly",
             use_random_seed=True,
-            specific_seed=TTS_SEED,
+            specific_seed=12345,
             api_name="/text_to_speech_app"
         )
+        return result[0] if isinstance(result, (list, tuple)) else result
+    except Exception as e:
+        print(f"Error in text to speech: {str(e)}")
+        return None
 
-# Create a single instance of EnglishTutor
-tutor = EnglishTutor()
+def chat_function(audio, history):
+    """الدالة الرئيسية للمحادثة"""
+    try:
+        # إذا لم يكن هناك صوت، نرجع بدون تغيير
+        if audio is None:
+            return history, None
 
-def format_response(response_dict):
-    """Format the response dictionary into a nice HTML string"""
-    html = f"<div style='font-size: 16px;'>"
-    html += f"<p>{response_dict['response']}</p>"
-    
-    if response_dict['corrections']:
-        html += f"<p><b>✍️ Corrections:</b> {response_dict['corrections']}</p>"
-    
-    if response_dict['vocabulary']:
-        html += f"<p><b>📚 Vocabulary:</b> {response_dict['vocabulary']}</p>"
-    
-    if response_dict['encouragement']:
-        html += f"<p><b>🌟 Encouragement:</b> {response_dict['encouragement']}</p>"
-    
-    html += "</div>"
-    return html
+        # تحويل الصوت إلى نص
+        user_message = convert_audio_to_text(audio)
+        if not user_message:
+            return history, None
 
-def chat(audio, history):
-    """Handle chat interactions"""
-    if audio is None:
-        # Return empty response if no audio
-        return history, None
-    
-    # Convert audio to WAV and transcribe
-    wav_path = convert_to_wav(audio)
-    if wav_path is None:
-        return history, None
+        # الحصول على رد البوت
+        bot_response = get_bot_response(user_message)
         
-    audio_text = transcribe_audio(wav_path)
-    # Clean up temporary file
-    os.remove(wav_path)
-    
-    if not audio_text:
-        return history, None
-    
-    # Get bot response
-    response = tutor.get_bot_response(audio_text)
-    
-    # Generate audio for the main response
-    audio_path = tutor.text_to_speech(response["response"])[0]
-    
-    # Format the complete response
-    formatted_response = format_response(response)
-    
-    # Update history in the correct format for gr.Chatbot
-    history = history or []
-    history.append((audio_text, formatted_response))
-    
-    return history, audio_path
+        # تحويل رد البوت إلى صوت
+        audio_response = text_to_speech(bot_response)
 
-def show_welcome():
-    """Show welcome message on startup"""
-    welcome = tutor.get_welcome_message()
-    audio_path = tutor.text_to_speech(welcome)[0]
-    return [(None, welcome)], audio_path
+        # تحديث المحادثة
+        history = history or []
+        history.append((user_message, bot_response))
+        
+        return history, audio_response
+    except Exception as e:
+        print(f"Error in chat function: {str(e)}")
+        return history, None
 
-# Create Gradio interface
-with gr.Blocks(css="footer {display: none}") as demo:
-    gr.Markdown("# 🤖 Sam - Your English Tutor")
-    gr.Markdown("Welcome to your personalized English learning session! Click the microphone and start speaking!")
+# إنشاء واجهة المستخدم
+with gr.Blocks() as demo:
+    gr.Markdown("# 🎓 English Tutor Chatbot")
     
-    chatbot = gr.Chatbot(
-        show_label=False,
-        height=400,
-        type="messages"
-    )
+    chatbot = gr.Chatbot(height=400)
     
     with gr.Row():
         audio_input = gr.Audio(
-            label="Speak here",
-            show_label=True,
-            type="filepath",
-            format="wav"
+            label="Your Voice",
+            type="filepath"
         )
         audio_output = gr.Audio(
-            label="Sam's Voice",
-            show_label=True,
+            label="Tutor's Voice",
             type="filepath"
         )
-    
-    # Handle audio input
+
+    # ربط الأحداث
     audio_input.change(
-        fn=chat,
+        chat_function,
         inputs=[audio_input, chatbot],
-        outputs=[chatbot, audio_output],
-        queue=False
-    )
-    
-    # Show welcome message on page load
-    demo.load_event(
-        fn=show_welcome,
-        inputs=None,
         outputs=[chatbot, audio_output]
     )
 
-# Launch the interface
+# تشغيل التطبيق
 if __name__ == "__main__":
-    demo.launch(
-    )
+    demo.launch(server_name="0.0.0.0", server_port=7860)