A 21-second look at a one-line brief becoming a captioned 9:16 short

A developer demonstrated a pipeline that converts a single-line text brief into a narrated, captioned vertical video, running the example brief "Synthwave drive through a neon city" with no API keys using mock scene art and an offline draft-quality voice. The system uses an LLM to generate a strictly validated list of scenes with image prompts, narration and timing, swaps image and voice providers through small adapters, renders steps in parallel via a job queue with retry and backoff, and assembles the output with ffmpeg using Ken Burns zooms, voice-synced scene durations and burned-in captions in 16:9, 9:16 or 1:1. The pipeline is offered as a paid kit ($49, Next.js + TypeScript + ffmpeg, 37 tests).

This short shows one real run of a pipeline that turns a single line of text into a narrated, captioned vertical video. The brief was "Synthwave drive through a neon city", run with zero API keys mock scene art and an offline, draft-quality voice . Here is what happens between the brief and the MP4, in plain terms: 1. An LLM writes the shots. The brief becomes a list of scenes, each with an image prompt, narration and timing. The output is validated strictly, so a broken JSON reply fails cleanly instead of crashing the render. 2. Providers make the pictures and the voice. Each provider is a small adapter, so you can swap a mock for a real image or voice service. 3. A job queue renders in parallel. Failed steps retry with backoff, and a crashed job can be recovered or retried with one click. 4. ffmpeg puts it together. Each scene gets a slow Ken Burns zoom, scenes stretch to fit the voice-over, and captions are burned in, timed to the narration. The output can be 16:9, 9:16 or 1:1. If you want the ffmpeg side in detail the 9:16 crop, the zoom filter and the caption timing , I wrote it up here: Building an AI Shorts pipeline with ffmpeg https://dev.to/syncoders/building-an-ai-shorts-pipeline-with-ffmpeg-916-ken-burns-zoom-captions-that-match-the-voice-1lc1 . Disclosure: the pipeline in the clip is our paid kit, the AI Video Pipeline Kit https://syncoder.gumroad.com/l/incack $49, full source, Next.js + TypeScript + ffmpeg, 37 tests .