Automated Faceless Video Generation: From Script to Short-Form Video

Chain LLM scriptwriting, ElevenLabs voice cloning, and Runway Gen-3 b-roll into an automated publishing workflow.

Prerequisites

ElevenLabs API
Runway or Luma API
FFmpeg installed

Step 1: Step 1: Storyboard Script Generation with Scene Breaks

Instruct the language model to output exact 4-second scene descriptions alongside matching voiceover audio lines.

Step 2: Step 2: Automated Audio Generation and Audio-Visual Assembly

Generate voice tracks with ElevenLabs, trigger image-to-video diffusion via Runway API, and splice together with FFmpeg.