
Doza Assist is a free, open source AI assistant designed specifically for editing documentary and spoken-word video content. Aimed at storytellers working with interviews, podcasts, testimonials, and other footage built from people talking, the tool focuses on helping editors identify and assemble the narrative moments that matter most. By analyzing locally dropped-in footage, Doza Assist finds key story beats, shapes them according to the user's established editing style, and exports rough cuts directly to major non-linear editing platforms such as Final Cut Pro, Premiere, and DaVinci Resolve.
The platform offers several core features tailored to narrative editing. Its AI Story Chat capability allows users to ask questions about the emotional arc, strongest hooks, or surprising moments within an interview, returning answers linked to specific, playable clips with timecodes. The Story Builder feature enables users to request narrative sequences, such as a pilot episode or a social cut, with each suggested clip accompanied by a title, timecode, and editorial note explaining its significance. Editors can rearrange, delete, or rebuild these sequences, and export them as pre-built timelines for further refinement in their preferred NLE.
A distinguishing aspect of Doza Assist is its ability to learn and adapt to an individual editor's style. By analyzing completed projects, the AI calibrates its suggestions to match the user's unique approach to storytelling—how they open scenes, build tension, and select moments to highlight or omit. This personalized model operates entirely on the user's Mac, ensuring that all footage, style profiles, and AI analysis remain local. No data is uploaded or shared, and there is no cloud processing involved, preserving privacy and control over creative work.
Doza Assist is distributed under the MIT license, requiring no subscription, trial, or account. It is positioned for a range of professionals, including those working in documentary film, corporate video, news, legal depositions, customer testimonials, and training content—any context where spoken-word footage needs to be distilled into meaningful stories. The tool is available as a local application for Mac and is built with a narrative-first approach, prioritizing understanding and assembling story arcs over simple keyword search.
In the Video editing space, Doza Assist takes a focused approach. It focuses on automating the process of finding story moments and assembling rough cuts in documentary editing. Doza Assist is an open-source project aimed at documentary editors and video storytellers. The project is open source (Open Source). The product ships for the web and macOS, and it can be self-hosted.
Doza Visuals builds and maintains Doza Assist, and the product first shipped in 2024. Among its 5 catalogued features are local transcription, story beat detection, and rough cut export.
Latest indexed changes and source events
Other apps tracked under the same category.