app.py

import warnings
from huggingface_hub import InferenceClient
import gradio as gr

warnings.filterwarnings('ignore')

# Initialize the language model
generator = InferenceClient("mistralai/Mixtral-8x7B-Instruct-v0.1")

def generate_script(host_name, listener_location, causes_climate_change, co2_level, effects_climate_change,
                    sea_level_rise, warming_rate, potential_solutions, individual_role, call_to_action,
                    TOPIC, DESCRIPTION):
    try:
        # Variables and template definitions...
        introduction_template = f"{host_name}, good morning! This is {listener_location}'s local radio station. Today we're talking about an issue that affects us all - {TOPIC}. It's a pressing issue that requires our immediate attention..."
        causes_template = f"The causes of {TOPIC} are {causes_climate_change}. Today, the level of CO2 in our atmosphere is {co2_level}, which is concerning..."
        effects_template = f"These activities result in {effects_climate_change}, leading to drastic changes in our environment. For instance, sea levels are rising at a rate of {sea_level_rise} per year, and global temperatures are increasing at a rate of {warming_rate} per decade..."
        solutions_template = f"But don't worry, there are solutions. {potential_solutions} are all steps we can take to mitigate these effects..."
        role_template = f"Each one of us plays a role in combating {TOPIC}. Even small actions can make a big difference. In fact, our location, {listener_location}, is particularly vulnerable to {TOPIC} due to its geographical features..."
        action_template = f"So, {listener_location}, why wait? Start taking steps today towards a greener future. Support local businesses that prioritize sustainability, reduce your carbon footprint, and voice your opinion to policy makers..."
        summary_template = f"In conclusion, {TOPIC} is a serious issue that requires our immediate attention. But by understanding its causes, effects, and potential solutions, we can all play a part in mitigating its impact. Thank you for joining us today, and remember, every small action counts!"

        # Combine templates based on the DESCRIPTION
        prompt_template = f"""{introduction_template} {causes_template} {effects_template} {solutions_template} {role_template} {action_template} {summary_template}
        
        TOPIC: {TOPIC}. DESCRIPTION: {DESCRIPTION}"""

        # Generate the script using the language model
        script = generator.text_generation(prompt_template)[0]['generated_text']
        if isinstance(response, list):
           script = response[0].get('generated_text', '')
        else:
           script = response.get('generated_text', '')
        # Split the script into sections
        sections = script.split("\n")

        # Calculate the word count for each section
        word_counts = [len(section.split()) for section in sections]

        # Check if any section exceeds the target word count
        for i, count in enumerate(word_counts):
            if count > 200:
                return f"Warning: Section {i + 1} exceeds the target word count. You may need to shorten this section."

        return script

    except Exception as e:
        error_message = f"Error: {e}"

        # Save error log to a file
        with open("./error_log.txt", "a") as log_file:
            log_file.write(error_message + "\n")

        return error_message

# Gradio interface setup...
iface = gr.Interface(fn=generate_script,
                   inputs=[gr.Textbox(label="Host Name", value="John"),
                           gr.Textbox(label="Listener Location", value="City"),
                           gr.Textbox(label="Causes Climate Change", value="human activities"),
                           gr.Number(label="CO2 Level", value=400),
                           gr.Textbox(label="Effects Climate Change", value="rising temperatures"),
                           gr.Number(label="Sea Level Rise", value=0.1),
                           gr.Number(label="Warming Rate", value=0.2),
                           gr.Textbox(label="Potential Solutions", value="renewable energy"),
                           gr.Textbox(label="Individual Role", value="reduce carbon footprint"),
                           gr.Textbox(label="Call To Action", value="act now")],
                   outputs="text")

# Launch the interface
iface.launch(debug=True)