Analysis updated 2026-05-18
Turn a novel or script into a structured shot breakdown for short drama production.
Manage character costumes, props, and visual consistency across video shots.
Batch-generate AI video clips and assemble them with dubbing and subtitles.
| xiakeman777/xiakeman-ai-short-drama | amarjitjim/browserpilot | boneskull/buggin | |
|---|---|---|---|
| Stars | 3 | 3 | 3 |
| Language | JavaScript | JavaScript | JavaScript |
| Last pushed | — | — | 2022-12-30 |
| Maintenance | — | — | Dormant |
| Setup difficulty | moderate | moderate | easy |
| Complexity | 3/5 | 3/5 | 2/5 |
| Audience | designer | developer | developer |
Figures from each repo's GitHub metadata at analysis time.
Docker self-host needs a connection to a local or cloud video-generation model endpoint, source code is not fully open.
Xiakeman is a creation workbench for producing AI-generated drama, short-form drama, and animated video content, built specifically around the Seedance2.0 video generation model. It helps creators take a story idea, a novel, or an annotated script and move it through the full production pipeline: structuring the script, managing characters and visual assets, designing shots, generating video, adding dialogue and subtitles, and assembling the final cut. The tool organizes this work into linked stages. Script organization converts novels or marked-up scripts into structured production materials. Character asset management tracks characters, scenes, costumes, props, and consistency references so the same character looks the same across shots. Shot design generates shot breakdowns, storyboards, and prompts for video generation. The video generation step connects to local or cloud-hosted video model services and supports running many clips as a batch. A dubbing and subtitles module handles dialogue lines, voice references, text-to-speech generation, subtitles, sound effects, and background music. Local final assembly of the finished clips uses FFmpeg. The project is available in three ways: an online hosted version for trying the product directly, a Docker-based self-hosted version for teams who want to run it on their own servers, and a Windows desktop application downloadable from the releases page. The repository itself mostly contains deployable runtime packages, Docker configuration, and setup documentation rather than the full editable source code. The software is proprietary. All rights are reserved by the author, and copying, modifying, redistributing, decompiling, reverse engineering, or creating derivative products from the release package is not permitted without written authorization. Commercial partnerships and the latest online version are pointed to on the project's own website. The primary language of the interface and documentation is Chinese.
A Chinese-language creation workbench for producing AI-generated short dramas and animated videos using the Seedance2.0 video model, from script to finished clip.
Mainly JavaScript. The stack also includes JavaScript, Docker, FFmpeg.
All rights reserved, copying, modifying, redistributing, or reverse engineering is not allowed without written permission.
Setup difficulty is rated moderate, with roughly 30min to a first successful run.
Mainly designer.
This repo across BitVibe Labs
Verify against the repo before relying on details.