linyqh/narratoai

★ 9,310PythonAudience · generalComplexity · 3/5LicenseSetup · moderate

Mindmap

mindmap
  root((NarratoAI))
    What it does
      Script generation
      Auto video cutting
      Voiceover synthesis
      Subtitle generation
    Tech stack
      Python
      Streamlit
      Docker
      AI LLM API
    Audience
      Content creators
      Drama reviewers
    Platforms
      Douyin
      Bilibili
      Short drama clips

mindmap root((NarratoAI)) What it does Script generation Auto video cutting Voiceover synthesis Subtitle generation Tech stack Python Streamlit Docker AI LLM API Audience Content creators Drama reviewers Platforms Douyin Bilibili Short drama clips

Click or tap to explore — scroll the page freely

Things people build with this

USE CASE 1

Automatically produce a narrated movie review video with voiceover and subtitles from raw film footage.

USE CASE 2

Generate short drama highlight clips with AI-written commentary for posting on Douyin or Bilibili.

USE CASE 3

Batch-produce narrated content without manually editing video or recording voiceovers by hand.

Tech stack

PythonStreamlitDocker

Getting it running

Difficulty · moderate Time to first run · 1h+

Requires an AI provider API key and at least 8 GB RAM, the interface and primary platform targets are Chinese-language.

Free for personal learning and research only, commercial use is not permitted.

In plain English

NarratoAI is a Python tool that automates the process of creating narrated video commentary, the kind of short-form content popular on Chinese video platforms where a voice-over explains the plot of a movie or TV drama over clips from it. The tool handles the full pipeline: it analyzes video footage, generates a script using an AI language model, cuts the video automatically, adds a synthesized voiceover, and generates subtitles, all without manual editing. The README is written primarily in Chinese and targets a Chinese-speaking audience. The tool is designed for content creators who make movie or drama review videos on platforms like Douyin (the Chinese version of TikTok) or Bilibili. It also supports a mode for mixing short drama clips. A cloud-hosted version is available at narratoai.cn for people who prefer not to install anything locally. To run it locally, you clone the repository and start it with Docker or by running a Python Streamlit web app. The interface opens in a browser. You connect it to an AI provider by adding an API key to a config file. The minimum recommended hardware is a 4-core CPU and 8 GB of RAM. A GPU is not required. The project notes that it is free and open source for personal learning and research only, not for commercial use. The README includes a warning that scammers have been repackaging and selling the software on social media under different names. The tool has been updated regularly since 2024, with recent releases adding speech cloning, support for additional text-to-speech engines, and frame-by-frame documentary analysis. It is built on concepts from earlier open-source video generation projects and extends them with the film narration workflow.

Copy-paste prompts

Prompt 1

Set up NarratoAI locally with Docker and walk me through creating my first narrated movie review video, what video format do I upload, which AI provider do I set, and how long does processing take?

Prompt 2

Show me how to configure NarratoAI to use a specific text-to-speech voice for the narration instead of the default, and how to adjust speaking speed.

Prompt 3

Explain NarratoAI's pipeline step by step: how does it decide which video clips to keep, how does it write the script, and how does it sync the voiceover timing to the cuts?

Prompt 4

I want to use NarratoAI on a Windows PC without Docker, what Python packages do I install, and are there any known compatibility issues to watch out for?

Prompt 5

How do I use NarratoAI's frame-by-frame documentary analysis mode for a nature documentary rather than a drama, what prompt or settings should I change?

Open on GitHub → Explain another repo

← linyqh on gitmyhub — every repo by this author, as a profile.

Verify against the repo before relying on details.