Update app.py

app.py

@@ -5,30 +5,38 @@ import gradio as gr
 # Load the OpenAI API key from environment variables (kept secret)
 openai.api_key = os.getenv("OPENAI_API_KEY")
 
-# Function to generate potential target audience groups using GPT-4o
-def generate_personas(company, industry, product, features_benefits):
-    # Constructing the prompt using the user inputs (company, industry, product, features, and benefits)
+# Predefined audience groups for the user to choose from
+predefined_audience_groups = [
+    "Tech-savvy Millennials (Aged 25-40, working in IT or creative industries)",
+    "Young Parents (Aged 30-45, seeking convenience in family products)",
+    "Retirees (Aged 60+, looking for comfort and health-focused products)",
+    "Environmentally Conscious Consumers (Aged 20-40, highly concerned about sustainability)",
+    "Luxury Shoppers (Aged 30-50, interested in premium brands and experiences)"
+]
+
+# Function to generate persona insights based on the chosen audience group and user input
+def generate_personas(audience_group, company, product, features_benefits):
     prompt = f"""
-    Generate three potential target audience groups for {company} in the {industry} industry. 
-    Focus on their product, {product}, which has features such as {features_benefits}.
+    Based on the selected audience group, {audience_group}, generate a detailed persona for the company {company}. 
+    Their product is {product}, which has features and benefits like {features_benefits}.
     """
     
-    # Using the OpenAI API with the GPT-4o engine to generate text based on the constructed prompt
+    # Call OpenAI's GPT-4o model to generate the persona insights
     response = openai.ChatCompletion.create(
         model="gpt-4o",  # Using GPT-4o for optimized performance
         messages=[{"role": "system", "content": "You are a helpful assistant."},
                   {"role": "user", "content": prompt}],
         max_tokens=200,
         n=1,
-        temperature=0.7  # Controls creativity (higher = more creative)
+        temperature=0.7
     )
     
-    # Return the generated text (target audience suggestions)
+    # Return the generated persona insights
     return response['choices'][0]['message']['content'].strip()
 
-# Function to handle further persona refinement via GPT-4o
+# Function to handle further persona refinement using GPT-4o
 def chat_with_persona_generator(persona_selection, prompt_selection):
-    # A dictionary mapping prompt options to specific requests for the persona refinement
+    # A dictionary mapping prompt options to specific requests for persona refinement
     prompt_map = {
         "Expand on the demographic attributes": "Provide detailed demographic attributes for our audience, including age, gender, location, and occupation.",
         "Elaborate on the key challenges": "What are the common challenges and pain points faced by the target audience in their daily life?",
@@ -42,10 +50,10 @@ def chat_with_persona_generator(persona_selection, prompt_selection):
         "Describe how our customers behave digitally": "How does the target audience behave digitally, including online shopping habits, browsing patterns, and social interactions?"
     }
 
-    # Construct the conversation prompt based on user selections
+    # Construct the prompt based on user selections
     prompt = f"{persona_selection}: {prompt_map.get(prompt_selection, 'Provide details about the target audience.')}"
     
-    # Using the GPT-4o model to generate a refined response based on the user prompt
+    # Using the GPT-4o model to generate a refined response based on the user's query
     response = openai.ChatCompletion.create(
         model="gpt-4o",  # Using GPT-4o for optimized performance
         messages=[{"role": "system", "content": "You are a helpful assistant."},
@@ -63,20 +71,21 @@ def persona_generator_interface():
     with gr.Blocks() as interface:
         gr.Markdown("## Welcome to the Customer Persona Generator!\nLet's get started with creating detailed personas for your marketing campaigns.")
         
+        # Dropdown to select predefined audience group
+        audience_group = gr.Dropdown(predefined_audience_groups, label="Select Predefined Audience Group")
+        
         # Input section to collect company and product details
         company = gr.Textbox(label="Company Name")
-        industry = gr.Textbox(label="Industry")
         product = gr.Textbox(label="Product")
         features_benefits = gr.Textbox(label="Features and Benefits")
         
-        # Button to generate the target audience groups
-        generate_btn = gr.Button("Generate Target Audience Groups")
+        # Button to generate persona insights based on the selected audience group
+        generate_btn = gr.Button("Generate Persona Insights")
         
-        # Output textbox to display generated audience groups
-        personas_output = gr.Textbox(label="Suggested Target Audience Groups", interactive=False)
+        # Output textbox to display generated persona insights
+        personas_output = gr.Textbox(label="Persona Insights", interactive=False)
         
-        # Dropdowns to select an audience group and a refinement prompt
-        persona_selection = gr.Dropdown(["Audience Group 1", "Audience Group 2", "Audience Group 3"], label="Select Audience Group")
+        # Dropdown to select an aspect of the persona for further refinement
         prompt_selection = gr.Dropdown([
             "Expand on the demographic attributes", 
             "Elaborate on the key challenges", 
@@ -88,17 +97,17 @@ def persona_generator_interface():
             "Suggest innovative communication channels", 
             "Analyze the factors contributing to purchasing decisions", 
             "Describe how our customers behave digitally"], 
-            label="Select Prompt")
+            label="Select Persona Aspect to Refine")
         
-        # Button to initiate persona refinement chat
-        chat_btn = gr.Button("Ask")
+        # Button to refine persona details
+        chat_btn = gr.Button("Refine Persona")
         
-        # Output textbox to display refined persona insights
-        chat_output = gr.Textbox(label="Persona Insights", interactive=False)
+        # Output textbox to display refined persona details
+        chat_output = gr.Textbox(label="Refined Persona Insights", interactive=False)
         
         # Actions when buttons are clicked
-        generate_btn.click(fn=generate_personas, inputs=[company, industry, product, features_benefits], outputs=personas_output)
-        chat_btn.click(fn=chat_with_persona_generator, inputs=[persona_selection, prompt_selection], outputs=chat_output)
+        generate_btn.click(fn=generate_personas, inputs=[audience_group, company, product, features_benefits], outputs=personas_output)
+        chat_btn.click(fn=chat_with_persona_generator, inputs=[audience_group, prompt_selection], outputs=chat_output)
 
     return interface