Create app.py

app.py

@@ -0,0 +1,159 @@
+import gradio as gr
+from groq import Groq
+from langchain_google_genai import ChatGoogleGenerativeAI
+import os
+import tempfile
+
+# -------------------- API Configuration --------------------
+# Set your API keys as environment variables in Hugging Face Spaces settings
+GROQ_API_KEY = os.environ.get("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", "")
+GOOGLE_API_KEY = os.environ.get("AIzaSyD2DMFgcL0kWTQYhii8wseSHY3BRGWSebk", "")
+
+# Initialize clients
+client = Groq(api_key=GROQ_API_KEY)
+llm = ChatGoogleGenerativeAI(
+    model="gemini-2.0-flash",
+    google_api_key=GOOGLE_API_KEY,
+    max_output_tokens=500
+)
+
+# -------------------- Core Functions --------------------
+def transcribe_audio(audio_path, language="ar"):
+    """Transcribe audio file using Groq Whisper"""
+    try:
+        with open(audio_path, "rb") as audio_file:
+            transcription = client.audio.transcriptions.create(
+                file=(os.path.basename(audio_path), audio_file.read()),
+                model="whisper-large-v3-turbo",
+                response_format="verbose_json",
+                language=language
+            )
+        return transcription.text, transcription.language
+    except Exception as e:
+        return f"Error in transcription: {str(e)}", None
+
+def get_ai_response(text):
+    """Get AI response from Gemini"""
+    try:
+        response = llm.invoke(text)
+        return response.content
+    except Exception as e:
+        return f"Error getting AI response: {str(e)}"
+
+def text_to_speech(text, language="ar"):
+    """Convert text to speech using Groq TTS"""
+    try:
+        # Select voice based on language
+        if language == "ar":
+            voice = "Amira-PlayAI"
+            model = "playai-tts-arabic"
+        else:
+            voice = "alloy"  # Default English voice
+            model = "playai-tts"
+        
+        response = client.audio.speech.create(
+            model=model,
+            voice=voice,
+            response_format="mp3",
+            input=text,
+        )
+        
+        # Save to temporary file
+        temp_file = tempfile.NamedTemporaryFile(delete=False, suffix=".mp3")
+        response.write_to_file(temp_file.name)
+        return temp_file.name
+    except Exception as e:
+        return None
+
+# -------------------- Gradio Interface Function --------------------
+def process_voice_chat(audio, language_choice):
+    """Main function to process voice input and generate response"""
+    
+    if audio is None:
+        return "Please provide an audio input", "", None
+    
+    # Map language choice to code
+    lang_map = {"Arabic": "ar", "English": "en"}
+    lang_code = lang_map.get(language_choice, "ar")
+    
+    # Step 1: Transcribe audio
+    transcription, detected_lang = transcribe_audio(audio, lang_code)
+    
+    if transcription.startswith("Error"):
+        return transcription, "", None
+    
+    # Step 2: Get AI response
+    ai_response = get_ai_response(transcription)
+    
+    if ai_response.startswith("Error"):
+        return transcription, ai_response, None
+    
+    # Step 3: Convert response to speech
+    # Use detected language if available, otherwise use selected language
+    output_lang = detected_lang if detected_lang else lang_code
+    audio_output = text_to_speech(ai_response, output_lang)
+    
+    return transcription, ai_response, audio_output
+
+# -------------------- Gradio Interface --------------------
+with gr.Blocks(title="Voice Chat Assistant", theme=gr.themes.Soft()) as demo:
+    gr.Markdown(
+        """
+        # 🎤 Voice Chat Assistant
+        Upload an audio file or record your voice to chat with AI. 
+        The AI will respond in the same language!
+        """
+    )
+    
+    with gr.Row():
+        with gr.Column():
+            language = gr.Radio(
+                choices=["Arabic", "English"],
+                value="Arabic",
+                label="Select Language"
+            )
+            audio_input = gr.Audio(
+                sources=["microphone", "upload"],
+                type="filepath",
+                label="Record or Upload Audio"
+            )
+            submit_btn = gr.Button("Process", variant="primary", size="lg")
+        
+        with gr.Column():
+            transcription_output = gr.Textbox(
+                label="Your Message (Transcription)",
+                lines=3
+            )
+            ai_response_output = gr.Textbox(
+                label="AI Response",
+                lines=5
+            )
+            audio_output = gr.Audio(
+                label="AI Voice Response",
+                type="filepath"
+            )
+    
+    # Button action
+    submit_btn.click(
+        fn=process_voice_chat,
+        inputs=[audio_input, language],
+        outputs=[transcription_output, ai_response_output, audio_output]
+    )
+    
+    gr.Markdown(
+        """
+        ### 📝 Instructions:
+        1. Select your language (Arabic or English)
+        2. Record your voice using the microphone or upload an audio file
+        3. Click "Process" to get AI response with voice output
+        
+        ### 🔑 Setup for Hugging Face Spaces:
+        Add these secrets in your Space settings:
+        - `GROQ_API_KEY`: Your Groq API key
+        - `GOOGLE_API_KEY`: Your Google API key
+        """
+    )
+
+# Launch the app
+if __name__ == "__main__":
+    demo.launch()