Update app.py

app.py

@@ -8,25 +8,18 @@ import os
 import speech_recognition as sr
 from pydub import AudioSegment
 
-# تحميل المتغيرات البيئية
 load_dotenv()
 
-# إعدادات API
 API_KEY = os.getenv("DEEPSEEK_API_KEY")
 HF_TOKEN = os.getenv("HUGGINGFACE_TOKEN")
 TTS_PASSWORD = os.getenv("TTS_PASSWORD")
 
-# التأكد من وجود المتغيرات المطلوبة
 if not all([API_KEY, HF_TOKEN, TTS_PASSWORD]):
     raise ValueError("Missing required environment variables!")
 
-# إعداد TTS
 TTS_CLIENT = Client("KindSynapse/Youssef-Ahmed-Private-Text-To-Speech-Unlimited", hf_token=HF_TOKEN)
-
-# إعداد محرك تحويل الكلام لنص
 recognizer = sr.Recognizer()
 
-# البرومبت الرئيسي
 MAIN_SYSTEM_PROMPT = {
     "role": "system",
     "content": """You are Sam, an intelligent and adaptive English tutor. Your responses must be in JSON format with these keys:
@@ -36,7 +29,6 @@ MAIN_SYSTEM_PROMPT = {
         'level_assessment': Current assessment of user's English level (beginner/intermediate/advanced),
         'encouragement': A motivating comment,
         'context_memory': Important details about the user to remember (interests, job, etc.)
-
         Guidelines:
         1. Maintain natural conversation while gathering user information
         2. Adapt language complexity to user's level
@@ -52,11 +44,9 @@ MAIN_SYSTEM_PROMPT = {
         - If user mentions hobbies → Engage in that topic
         - For beginners → Use simple words and short sentences
         - For intermediate/advanced → Introduce more complex vocabulary
-
         Always maintain conversation history and adapt based on it."""
 }
 
-# برومبت الترحيب
 WELCOME_PROMPT = {
     "role": "system",
     "content": """Create a warm welcome message that:
@@ -80,10 +70,8 @@ class EnglishTutor:
             "profession": None,
             "goals": None
         }
-        self.is_first_interaction = True
 
     def get_welcome_message(self):
-        """توليد رسالة ترحيب"""
         try:
             response = requests.post(
                 "https://api.deepseek.com/v1/chat/completions",
@@ -102,9 +90,7 @@ class EnglishTutor:
             return "Hi! I'm Sam, your English tutor. What's your name?"
 
     def get_bot_response(self, user_message):
-        """الحصول على رد من البوت"""
         try:
-            # إضافة رسالة المستخدم للمحادثة
             self.chat_history.append({"role": "user", "content": user_message})
             
             response = requests.post(
@@ -120,16 +106,13 @@ class EnglishTutor:
             
             bot_response = json.loads(response.json()["choices"][0]["message"]["content"])
             
-            # تحديث معلومات المستخدم
             if "level_assessment" in bot_response:
                 self.user_info["level"] = bot_response["level_assessment"]
             if "context_memory" in bot_response:
                 self._update_user_info(bot_response["context_memory"])
             
-            # تنسيق الرد
             formatted_response = self._format_response(bot_response)
             
-            # إضافة رد البوت للمحادثة
             self.chat_history.append({"role": "assistant", "content": json.dumps(bot_response)})
             
             return formatted_response
@@ -138,14 +121,12 @@ class EnglishTutor:
             return "I apologize, but I couldn't process that properly. Could you try again?"
 
     def _update_user_info(self, context_memory):
-        """تحديث معلومات المستخدم من الذاكرة السياقية"""
         if isinstance(context_memory, dict):
             for key in self.user_info:
                 if key in context_memory:
                     self.user_info[key] = context_memory[key]
 
     def _format_response(self, response_dict):
-        """تنسيق الرد بشكل جميل"""
         formatted = response_dict["response"]
         
         if response_dict.get("corrections"):
@@ -160,9 +141,7 @@ class EnglishTutor:
         return formatted
 
 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'
@@ -178,7 +157,6 @@ def convert_audio_to_text(audio_path):
         return None
 
 def text_to_speech(text):
-    """تحويل النص إلى صوت"""
     try:
         result = TTS_CLIENT.predict(
             password=TTS_PASSWORD,
@@ -194,72 +172,96 @@ def text_to_speech(text):
         print(f"Error in text to speech: {str(e)}")
         return None
 
-# إنشاء كائن المعلم
 tutor = EnglishTutor()
 
-def chat_function(audio, history):
-    """الدالة الرئيسية للمحادثة"""
+def initialize_chat():
     try:
-        # إذا كانت أول محادثة، نعرض رسالة الترحيب
-        if not history:
-            welcome = tutor.get_welcome_message()
-            welcome_audio = text_to_speech(welcome)
-            return [(None, welcome)], welcome_audio
+        welcome = tutor.get_welcome_message()
+        welcome_audio = text_to_speech(welcome)
+        return [(None, welcome)], welcome_audio, ""
+    except Exception as e:
+        print(f"Error initializing chat: {str(e)}")
+        return [(None, "Hi! I'm Sam, your English tutor. What's your name?")], None, ""
 
-        # إذا لم يكن هناك صوت، نرجع بدون تغيير
+def process_audio(audio, history, transcript):
+    try:
         if audio is None:
-            return history, None
+            return history, None, transcript
 
-        # تحويل الصوت إلى نص
         user_message = convert_audio_to_text(audio)
         if not user_message:
-            return history, None
+            return history, None, transcript
 
-        # الحصول على رد البوت
         bot_response = tutor.get_bot_response(user_message)
-        
-        # تحويل رد البوت إلى صوت
         audio_response = text_to_speech(bot_response)
 
-        # تحديث المحادثة
         history = history or []
         history.append((user_message, bot_response))
         
-        return history, audio_response
+        new_transcript = transcript + f"\n\n🎤 You said: {user_message}\n🤖 Sam replied: {bot_response}"
+        
+        return history, audio_response, new_transcript
     except Exception as e:
-        print(f"Error in chat function: {str(e)}")
-        return history, None
+        print(f"Error in process_audio: {str(e)}")
+        return history, None, transcript
+
+def clear_chat():
+    global tutor
+    tutor = EnglishTutor()
+    return initialize_chat()
 
-# إنشاء واجهة المستخدم
 with gr.Blocks(theme=gr.themes.Soft()) as demo:
     gr.Markdown("# 🎓 English Learning Assistant")
     gr.Markdown("Speak naturally and I'll help you improve your English!")
     
-    chatbot = gr.Chatbot(
-        height=400,
-        bubble_full_width=False,
-        show_label=False
-    )
+    with gr.Row():
+        with gr.Column(scale=2):
+            chatbot = gr.Chatbot(
+                height=400,
+                bubble_full_width=False,
+                show_label=False
+            )
+            
+            with gr.Row():
+                audio_input = gr.Audio(
+                    label="Your Voice",
+                    type="filepath",
+                    show_label=True
+                )
+                audio_output = gr.Audio(
+                    label="Tutor's Voice",
+                    type="filepath",
+                    show_label=True
+                )
+        
+        with gr.Column(scale=1):
+            gr.Markdown("### 📝 Conversation Transcript")
+            transcript_display = gr.Textbox(
+                lines=15,
+                max_lines=15,
+                show_label=False,
+                interactive=False,
+                placeholder="Your conversation transcript will appear here..."
+            )
     
     with gr.Row():
-        audio_input = gr.Audio(
-            label="Your Voice",
-            type="filepath",
-            show_label=True
-        )
-        audio_output = gr.Audio(
-            label="Tutor's Voice",
-            type="filepath",
-            show_label=True
-        )
-
-    # ربط الأحداث
+        clear_btn = gr.Button("🔄 Clear Chat", variant="secondary")
+    
     audio_input.change(
-        chat_function,
-        inputs=[audio_input, chatbot],
-        outputs=[chatbot, audio_output]
+        process_audio,
+        inputs=[audio_input, chatbot, transcript_display],
+        outputs=[chatbot, audio_output, transcript_display]
+    )
+    
+    clear_btn.click(
+        clear_chat,
+        outputs=[chatbot, audio_output, transcript_display]
+    )
+    
+    demo.load(
+        initialize_chat,
+        outputs=[chatbot, audio_output, transcript_display]
     )
 
-# تشغيل التطبيق
 if __name__ == "__main__":
     demo.launch()