Claude Code turned one long video take into a 15-second ad
A creator demonstrates an AI-powered video editing workflow that automatically transformed a single 5-minute phone recording into a polished 15-second vertical ad. The system used silence detection, speech recognition, facial tracking, and automated shot composition to generate cuts, captions, and graphics without manual timeline editing.
Summary
The creator explains their need to produce personal face-to-camera content rather than rely on user-generated content, as their presence performs better for advertising. To accomplish this efficiently, they filmed a single continuous 5-minute take on their phone while in a co-working space, then transferred the footage to a MacBook via AirDrop. They then used an automated AI system that performed multiple sophisticated editing tasks sequentially: silence detection identified where the speaker was talking, Whisper (a speech recognition model) provided precise word timing to sharpen edit points, and facial tracking cropped and followed the creator's face throughout the video. The system automatically selected eight cuts from a TSP (Time Series Prediction) file and assigned shot sizes—wide, medium, and close-up—to each cut. The creator then prompted the system to find their branding information and graphics, open a simulator, capture an app screenshot, and composite it into the video. Finally, the system generated animated captions displaying one word at a time across all 57 words in the video, styled with white text and black outlines. The result was a complete 15-second vertical advertisement ready for platform delivery, created entirely without traditional video editing—no timeline, no project file, no manual editor interaction.
Key Insights
- The creator determined that personal face-to-camera UGC content performs better than generic user-generated content, driving their need to produce their own videos
- Whisper's word timing capability serves as the foundation for precise edit point creation by sharpening each edge where speech pauses occur
- Facial tracking automation enables dynamic shot size variation across eight cuts, automatically assigning wide, medium, and close framing without manual selection
- The system prompted for branding information and graphics, then independently located simulator screenshots and composited them into the video in proper context
- A complete 15-second polished advertisement was generated from a single 5-minute phone recording without any traditional video editing tools, timeline, or project files
Topics
Transcript
[0:00] I need to do my own videos because UGC content doesn't do as well as this face. So, I needed long-form to then be able to upload and cut up. I was in my a co-work space and filmed it with my phone, AirDropped it to the MacBook, and then this is what happened. One phone take, 5 minutes of me talking, find the speech. So, silence detection marks where she's talking. Word timing from Whisper sharpens each edge, and then it tracked my face. So, the crop follows her, and each cut gets a shot size, wide, mid, and close. Then place the [0:30] cuts, so eight of them picked from its own TSP file. I went and…
Full transcript available for MurmurCast members
Sign Up to AccessMore from How I AI
What are channels in Claude Duet?
Claude Duet features a channels system inspired by Slack that allows users to organize conversations into separate subthreads within a main chat. This enables more structured collaboration by creating dedicated channels for specific projects, such as a sales channel for sales-related work.
Claude Duet needs multiplayer (please!)
The speaker argues that Claude and similar AI products are currently optimized for single-player use and lack proper multiplayer capabilities. Despite integrations like Claude tags and Slack, the speaker contends that sharing agents with their small team of five people remains problematic, preventing these tools from becoming part of their regular workflow.
Claude Duet is great, but SLOW
The speaker expresses frustration with the slowness of AI model Claude, which impacts their willingness to engage with it. They compare its responsiveness unfavorably against more responsive alternatives, despite acknowledging Claude's capabilities.
Claude Duet should be more annoying
The speaker emphasizes the need for agents like Claude Duet to be more engaging and proactive to avoid being forgotten in the competitive landscape of digital assistants. They critique the current methods of communication, such as morning briefings, as ineffective in attracting user interest.
A perfect TikTok cut from a one sentence prompt
Claude code automatically transforms a single sentence prompt into a polished, TikTok-ready short-form video with captions, jump cuts, and zooms in approximately 4 seconds with no manual editing required. The tool demonstrates AI's capability to understand and execute high-quality short-form video production autonomously.