How to Create a VIRAL Talking Baby Podcast with AI (No-Code n8n Tutorial).

Key Concepts:

• No-code AI Automation: Using platforms like Naden to automate tasks without extensive coding.
• AI Model Integration: Combining different AI models (OpenAI, 11 Labs, Hedra) for content creation.
• API Keys: Authentication keys required to access and use AI services.
• JSON Workflows: Blueprints in Naden that define the steps in an automated process.
• HTTP Requests: Sending requests to external APIs to retrieve or send data.
• Form Submissions: Using forms to collect user input and trigger automated workflows.
• Image and Audio Generation: Using AI to create images and audio content based on prompts.
• Lip-Sync Video Generation: Combining audio and images to create videos with synchronized lip movements.
• Credit-Based Pricing: Cost structure where AI services charge based on credits consumed per operation.

1. Overview of the Automation

The video demonstrates how to create automated baby-style podcasts using a combination of AI tools and the Naden no-code platform. The process involves generating audio and images of babies discussing various topics and then combining them into short videos suitable for platforms like TikTok, Instagram, and YouTube. The presenter highlights two methods: a "lazy way" using a pre-built blueprint and a step-by-step approach for building the automation from scratch.

2. Lazy Method: Using a Pre-Built Blueprint

• Naden Community: The presenter emphasizes joining the Naden community to access pre-built blueprints.
• Blueprint Download: Users can download a JSON workflow (blueprint) from the community and import it into their Naden instance.
• Form Submission: The blueprint includes a form where users can specify the baby's ethnicity, hairstyle, and the topic of conversation.
• Automated Process: Once the form is submitted, the workflow automatically generates audio using 11 Labs, creates an image using OpenAI's GPT image model, and combines them into a video using Hedra.
• API Key Configuration: The only required modification is updating the API keys for OpenAI, 11 Labs, and Hedra within the blueprint.

3. Step-by-Step Method: Building the Automation from Scratch

• Form Creation: The first step is creating a form in Naden to collect user input for the baby's ethnicity, hairstyle, and topic of discussion.
• OpenAI Prompt Generation: An OpenAI node is used to generate a prompt for the image based on the user's input. The prompt is designed to create a high-resolution image of a cute baby with specific characteristics, wearing headphones, and speaking into a podcast microphone.
• Image Generation with OpenAI: An HTTP request node sends the generated prompt to OpenAI's image API to create the image. The presenter uses the GPT image-1 model and specifies the image size as 1024x1536 pixels.
• Image Conversion: The image data, which is initially in Base64 format, is converted into a file using a "Convert JSON to Binary File" node.
• Hedra Integration: The generated image is uploaded to Hedra using an HTTP request node. This requires setting up custom authentication with Hedra's API key.
• Audio Generation with 11 Labs: Similar to image generation, an OpenAI node is used to generate a prompt for the audio. The prompt is designed to create a podcast-style monologue in the style of Joe Rogan, discussing the specified topic.
• Audio Generation with 11 Labs: An HTTP request node sends the generated prompt to 11 Labs to create the audio file.
• Audio Upload to Hedra: The generated audio file is uploaded to Hedra using the same process as the image upload.
• Merging Assets: A merge node combines the image and audio metadata.
• Splitting Metadata: A code node splits the metadata to extract the necessary IDs for the image and audio assets.
• Video Generation with Hedra: An HTTP request node sends the image and audio IDs to Hedra's video generation API to create the final video.
• Video Retrieval and Download: The generated video is retrieved from Hedra using another HTTP request node and downloaded as a binary file.
• Google Drive Upload and Conversion: The video is uploaded to Google Drive to convert it to MP4 format.
• YouTube Upload: Finally, the MP4 video is uploaded to YouTube using the YouTube API.

4. Cost Analysis

• Hedra Pricing: Hedra charges 3.5 to 7 credits per second of video generated. A basic plan costs $10 per month and provides 1,000 credits.
• OpenAI Pricing: OpenAI's image generation costs vary depending on the image size.
• Overall Cost: The presenter estimates that a 60-second video could cost around $1, including Hedra's platform fee and OpenAI's image generation costs.

5. Technical Details and API Integrations

• Naden: A no-code/low-code AI platform used to orchestrate the entire workflow.
• OpenAI: Used for generating both image and audio prompts, as well as generating the image itself. The GPT-4 model is recommended for high-quality image generation.
• 11 Labs: Used for generating the audio based on the generated prompt.
• Hedra: Used for combining the image and audio into a lip-sync video. The Hedra Character 3 model is used for video generation.
• Google Drive: Used for converting the video to MP4 format.
• YouTube API: Used for uploading the final video to YouTube.

6. Community Resources

• Naden Community: Provides access to pre-built blueprints and support.
• API Documentation: The presenter refers to the API documentation for OpenAI, 11 Labs, and Hedra for detailed information on API endpoints, authentication, and request parameters.

7. Notable Quotes

• "It's wild man because we're wired for survival not passive income." (Opening statement setting the context of automation)
• "This is going to be a total automation i'm going to show you how to generate this all you have to do is select the topic of discussion the ethnicity of the baby and you have to press a button and this will generate." (Describing the ease of use of the automated process)
• "The reason why I'm using the OpenAI image generation is because again this is probably the best um model in the market right now." (Justifying the choice of OpenAI for image generation)

8. Conclusion

The video provides a comprehensive guide to automating the creation of baby-style podcasts using AI and no-code tools. It covers both a quick, "lazy" method using pre-built blueprints and a detailed, step-by-step method for building the automation from scratch. The presenter emphasizes the importance of integrating different AI models, managing API keys, and understanding the cost implications of using these services. The final product is a fully automated system that can generate and upload videos to YouTube with minimal user intervention.