title	PodcastLab
emoji	🎙️
colorFrom	blue
colorTo	purple
sdk	streamlit
pinned	false
license	mit
short_description	AI podcast creator using Google's Imagen, Gemini, and Veo.

PodcastLab - AI-Powered Podcast Creator

🎙️ Create professional podcasts with AI using Google's Imagen, Gemini, and Veo models

Demo

Untitled.video.-.Made.with.Clipchamp.mp4

Features

• AI-Generated Content: Create complete, end-to-end podcast scripts using Google's Gemini AI
• Google Search Grounding: Access real-time, factual information with Google Search integration
• Visual Storytelling: Generate custom podcast cover images with Google's Imagen
• Video Podcasts: Create square video podcasts using Google's Veo model
• Professional Quality: High-quality audio and video output
• Easy to Use: Simple Streamlit interface for seamless podcast creation
• Source Citations: Automatic citation generation for factual claims

How to Use

1. Enter Your Topic: Input any topic you want to create a podcast about
2. Enable Search Grounding: Toggle the "Use Google Search for factual data" option for current information
3. Configure Settings: Choose the number of parts, duration, and other preferences
4. Generate Content: Let AI create the script, images, and video content
5. Download Results: Get your complete podcast package ready for publishing

Google Search Grounding

The podcast agent now supports Google Search grounding, which allows Gemini to access real-time web information for more accurate and up-to-date content. When enabled:

• Real-time Information: Get current facts and data about your topic
• Source Citations: Automatic generation of citations for factual claims
• Search Queries: See what search queries were used to gather information
• Source Links: Access the original sources used in the podcast script

This is particularly useful for topics involving:

• Current events and news
• Recent technological developments
• Up-to-date statistics and data
• Latest trends and developments

Script Structure

Each podcast episode is generated as 3 complete segments:

1. Part 1 (Introduction): Welcome message, topic introduction, and what listeners will learn
2. Part 2 (Main Content): Key insights, facts, and detailed discussion with host interactions
3. Part 3 (Conclusion): Key takeaways, wrap-up, and call-to-action

Each part is designed to feel like a complete segment of a real podcast episode with natural host conversations and smooth transitions.

Requirements

• Google AI API Key (set in Streamlit Cloud secrets or as GOOGLE_AI_API_KEY environment variable)
• Internet connection for AI model access

Deployment on Streamlit Cloud

Option 1: Using Streamlit Cloud Secrets (Recommended)

1. Fork this repository to your GitHub account
2. Go to Streamlit Cloud
3. Click "New app" and connect your GitHub repository
4. Set up secrets in the Streamlit Cloud dashboard:
   • Go to your app's settings
   • Navigate to "Secrets" section
   • Add your Google AI API key:

     [secrets]
     GOOGLE_AI_API_KEY = "your-google-ai-api-key-here"

5. Deploy your app

Option 2: Using Environment Variables

1. Fork this repository to your GitHub account
2. Go to Streamlit Cloud
3. Click "New app" and connect your GitHub repository
4. Set environment variables in the Streamlit Cloud dashboard:
   • Go to your app's settings
   • Navigate to "Environment variables" section
   • Add: GOOGLE_AI_API_KEY = your-google-ai-api-key-here
5. Deploy your app

Getting Your API Key

1. Visit Google AI Studio
2. Sign in with your Google account
3. Navigate to the API section
4. Create a new API key
5. Copy the key and add it to your Streamlit Cloud secrets or environment variables

Technical Details

• Framework: Streamlit
• AI Models: Google Gemini, Imagen, Veo
• Video Processing: MoviePy
• Image Processing: Pillow
• Aspect Ratio: Square (1:1) for videos
• Deployment: Streamlit Cloud

License

MIT License - feel free to use and modify for your projects!

---

Note: This application requires a valid Google AI API key to function. The key is not included in the repository for security reasons. The cost may range between $5 to $9 per video creation. Use it at your own risk.