Update app.py

app.py

@@ -54,93 +54,86 @@ import gradio as gr
 from huggingface_hub import InferenceClient
 import tempfile
 
-# Initialize clients for both chat and speech-to-text
+# Initialize clients
 chat_client = InferenceClient("Qwen/Qwen2.5-Coder-32B-Instruct")
-stt_client = InferenceClient("openai/whisper-large-v3")  # Using OpenAI's Whisper model for STT
+stt_client = InferenceClient("openai/whisper-large-v3")
 
 def transcribe_audio(audio_file):
     """Convert audio to text using Whisper model"""
     with open(audio_file, "rb") as f:
         return stt_client.automatic_speech_recognition(f.read())
 
-def respond(message, history: list[tuple[str, str]], audio=None):
-    # If audio input is provided, transcribe it first
-    if audio:
-        message = transcribe_audio(audio)
-    
-    system_message = "You are a friendly Chatbot. If the user query is product-related, provide structured product recommendations based on intent and relevance."
-    max_tokens = 2048
-    temperature = 0.7
-    top_p = 0.95
-
+def respond(history, query):
+    system_message = "You are a friendly Chatbot. Provide structured product recommendations based on user queries."
     messages = [{"role": "system", "content": system_message}]
 
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+    # Build conversation history
+    for entry in history:
+        messages.append({"role": "user", "content": entry[0]})
+        if entry[1]:  # Only add assistant response if present
+            messages.append({"role": "assistant", "content": entry[1]})
 
-    product_prompt = ("Given a user's search query, recommend relevant products...")
-    messages.append({"role": "user", "content": f"{product_prompt}\n\nUser query: {message}"})
+    # Add product recommendation prompt
+    product_prompt = ("Analyze this query and provide recommendations: ")
+    messages.append({"role": "user", "content": f"{product_prompt}\n{query}"})
 
+    # Generate streamed response
     response = ""
     for chunk in chat_client.chat_completion(
         messages,
-        max_tokens=max_tokens,
+        max_tokens=2048,
         stream=True,
-        temperature=temperature,
-        top_p=top_p,
+        temperature=0.7,
+        top_p=0.95,
     ):
         token = chunk.choices[0].delta.content
         response += token
-        yield response
+        history[-1] = (query, response)  # Update last history entry
+        yield history
 
-# Custom CSS for better visual appearance
+# Custom CSS for better styling
 css = """
-.gradio-container {
-    background: linear-gradient(135deg, #f5f7fa 0%, #c3cfe2 100%);
-    min-height: 100vh;
-}
-.audio-input {
-    background: white !important;
-    border-radius: 10px !important;
-    padding: 20px !important;
-}
+.gradio-container { background: #f0f4f8 !important; }
+.audio-input { background: white !important; border-radius: 10px; }
 """
 
 with gr.Blocks(css=css) as demo:
-    gr.Markdown("# Smart Product Assistant 🎤🛒")
+    gr.Markdown("# Smart Product Assistant 🎤🛍️")
+    
     with gr.Row():
-        with gr.Column(scale=2):
-            chatbot = gr.Chatbot(height=600)
-        with gr.Column(scale=1):
-            gr.Markdown("## Input Methods")
+        chatbot = gr.Chatbot(height=600)
+        
+        with gr.Column():
             with gr.Tab("Text Input"):
-                text_input = gr.Textbox(placeholder="Type your query here...", label="Text Input")
+                text_input = gr.Textbox(label="Type your query")
             with gr.Tab("Voice Input"):
                 audio_input = gr.Audio(
                     sources="microphone",
                     type="filepath",
-                    label="Speak your query",
+                    label="Record your query",
                     elem_classes="audio-input"
                 )
             submit_btn = gr.Button("Submit", variant="primary")
 
-    def process_input(text, audio, history):
-        if audio:
-            return audio, history
-        elif text:
-            return text, history
-        return "", history
+    def process_inputs(text, audio, history):
+        """Handle both text and audio inputs"""
+        query = text.strip()
+        if audio and not query:
+            query = transcribe_audio(audio)
+        
+        if query:
+            # Add new entry to history with empty response
+            return history + [[query, None]]
+        return history
 
     submit_btn.click(
-        process_input,
+        process_inputs,
         [text_input, audio_input, chatbot],
-        [text_input, chatbot]
+        chatbot,
+        queue=False
     ).then(
         respond,
-        [text_input, chatbot, audio_input],
+        [chatbot, text_input],
         chatbot
     )