Substack Re-Stack

This first embedded link - is the NotebookLM podcast highlighting takeaways from this past weeks Substack articles - Re-Stacked!

This past week’s Substack articles describe AI as a collaborative tool for creative professionals, emphasizing its role in enhancing creative processes rather than replacing human creativity.

AI should be treated as a "robotic employee" awaiting instruction, not just a tool for data entry.

AI can provide valuable assistance in various creative tasks. You can use ChatGPT with GPT Vision to:

• Generate recipes from a photo of ingredients.

• Solve visual puzzles.

• Write scripts, including dialogue.
• Create character stills and animations based on scripts.

"The Human Element" remains crucial in guiding and refining AI's output. You must direct the AI and leverage your intuition and observational skills to achieve the desired outcome. For instance, in the puzzle-solving example, ChatGPT required specific instructions based on human observation to arrive at the solution.

The articles provide specific guidance on how AI can be used to improve copywriting and scriptwriting:

• Copywriting:Focus on using vivid language that evokes strong imagery. Your prompts should be clear and descriptive enough to create a visual in the reader's mind.
• Use sensory details and literary devices to enhance the impact of your writing. Incorporate details related to sight, sound, taste, touch, and smell to create a richer experience for the reader.
• Ensure your copy is specific, verifiable, and unique. Follow the three rules of good copywriting: Can I visualize it? Can I falsify it? Can nobody else say this?
• Scriptwriting:Study the techniques of renowned scriptwriters like Aaron Sorkin and David Mamet. Analyze their work to understand how they craft natural dialogue, create tension, and use language to reveal character.
• Provide detailed context and background information to the AI when prompting for a script. This helps the AI understand the characters, relationships, and overall tone of the scene.
• Strive for a natural cadence and avoid excessive framing or context within the dialogue. Ensure the conversation flows organically and sounds like a genuine interaction between humans.

The articles detail a process for generating animated dialogue scenes using a combination of AI tools and human input:

• Use Midjourney to generate still images of your characters. Craft detailed prompts that describe the character's appearance, including specific details like clothing and setting.

• Record yourself acting out the dialogue script, capturing both audio and video. Use a tool like Camtasia to record yourself reading the lines for each character. Aim for a neutral background to avoid distractions in the final animation.

• Utilize Runway's Gen-3 Act-One to animate your characters based on your recorded performance. Upload the still images of your characters and the video of your acting to Runway. The tool will generate a video that maps your movements and facial expressions onto the character images.
• Replace your voice with AI-generated voices using ElevenLabs. Choose voices that match the characters' personalities. Sync the new audio with the animated footage in a video editing software like Adobe Premiere.
• Add a sound bed from Udio to enhance the atmosphere of the scene. Choose a music style that fits the mood of the dialogue. Use audio ducking to ensure the dialogue remains clear over the music.

• Consider creating multiple views of your characters to add visual variety to the animation. Use Midjourney to generate images of the same character from different angles. Run these through Runway's Gen-3 Act-One to create more dynamic scenes.

The articles highlight the potential of AI to significantly streamline the animation process while maintaining a high level of quality and creative control.