zcbacxc/movie-narrator

🎬 Generate narrated movie recap videos from a single prompt.

What it solves

Movie Narrator automates the creation of movie recap videos. It eliminates the manual effort of writing scripts, recording voiceovers, generating subtitles, and editing clips together, allowing users to produce a fully narrated video from a single prompt.

How it works

The project uses a 16-step sequential pipeline to transform a movie title into a finished video:

  1. Content Generation: It uses LLMs to research the plot (via TMDB) and write a recap script based on a chosen perspective (e.g., omniscient or detective).
  2. Audio & Subtitles: It converts the script to speech using TTS (Edge-TTS by default) and generates synchronized SRT subtitles, which can be translated into multiple languages.
  3. Visual Assembly: It detects scenes in the source video, matches them to the script segments, and integrates background music.
  4. Rendering: The final assets are rendered into a 1080p or 4K video in various formats (e.g., 16:9 or 9:16).

Who it’s for

It is designed for content creators who make movie recap videos for social media or platforms like YouTube and TikTok, as well as developers looking for an extensible framework for automated video narration.

Highlights

  • Multi-candidate "Horse Race": Generates multiple script variations and automatically selects the best one.
  • Reference Imitation: Can extract the narration style from a viral reference video to apply it to a new movie.
  • VLM Integration: Uses Vision Language Models (VLM) for scene captioning via cloud APIs.
  • Flexible Infrastructure: Supports various OpenAI-compatible LLMs (Ollama, Zhipu, etc.) and multiple TTS providers.
  • QA Gate: Includes automated detection for black frames or slideshow-like risks in the final output.

Related

  • Project
    AceDataCloud/Nexior

    Nexior is an MIT‑licensed, Vue‑based app that aggregates 40+ chat, image, audio and video AI models into a single self‑hostable UI. It offers one‑click Vercel or Docker deployment, BYOK or a single AceData key, and includes built‑in user accounts, payments and referral mechanics, making it a ready‑made SaaS starter for AI products.

  • Project
    xeronsh/openOii

    An AI comic-video generation project that uses LangGraph to orchestrate a multi-agent pipeline from story ideas to final video synthesis.

  • Project
    deepmodeling/dpgen

    DP‑GEN is a Python platform that automates the creation of deep‑learning interatomic potentials (Deep Potential models). It runs a concurrent learning loop—sampling structures, selecting the most informative ones for DFT calculations, and retraining the model—while handling job scheduling on HPC clusters and interfacing with many MD and quantum‑chemistry codes.

  • Project
    Wakals/CoVT

    CoVT is a framework that enables Vision-Language Models to reason using continuous visual tokens, improving their spatial reasoning and geometric awareness by grounding semantic thoughts in perceptual cues.