Podcast episodes
Audio-first, production-ready. Brief, script, voice, assembly, published.
Audio-first is not a limitation — it is the design. Every podcast episode you produce is already a video waiting for a visual layer.
The insight
The pipeline runs the same way for both formats: brief → script → voice → assembly → output. When you produce a podcast episode, the visual layer is null. When you are ready for video, the visual layer is your locked character composited onto the audio track that already exists.
Live proof
This is not a hypothetical pipeline. Stig and Bea — our own AI-hosted podcast — is produced on this infrastructure. Every episode: brief, script, voice, assembly, published.
Listen on Spotify →What Video produces
Podcast episodes
Audio-first, production-ready. Brief, script, voice, assembly, published.
Short-form video clips
Locked character composited onto the audio track that already exists.
Full video at V3
Storyboard, voice, and character — composed scene by scene.
Scene-level assets for ads
Individual scenes flow forward into Ads as variant inputs.
Production flow
Brief or article comes in from Docs.
Script generated and presented for approval in storyboard UI.
Review scene by scene — live TTS playback, word-level highlighting, inline edits.
Individual scenes re-renderable without re-running the full pipeline.
Final assembly via audio engine.
Output goes to Publish and Ads.
How it connects
Video reads your character spec from Image and scripts from Docs. Outputs feed Publish for distribution and Ads for variant assembly. The episode that goes to your podcast feed and the clip that runs as an ad come from the same source file.
Same locked price. Every tool, as it ships.
Join as a founding member →