MCPcopy Index your code
hub / github.com/linyqh/NarratoAI

github.com/linyqh/NarratoAI @v0.8.4 sqlite

repository ↗ · DeepWiki ↗ · release v0.8.4 ↗
1,498 symbols 5,329 edges 137 files 645 documented · 43%
README

NarratoAI 😎📽️

An all-in-one AI-powered tool for film commentary and automated video editing.🎬🎞️

📖 English | 简体中文 | 日本語

NarratoAI is an automated video narration tool that provides an all-in-one solution for script writing, automated video editing, voice-over, and subtitle generation, powered by LLM to enhance efficient content creation.

🔥 Highly Recommended: The new paradigm of VibeCut —— Speclip , a true editing Agent! 👉 Click to download for free

madewithlove GitHub license GitHub issues GitHub stars

💬 Join the open source community to get project updates and the latest news.

🎉🎉🎉 Official Documentation 🎉🎉🎉

Home

Video Review Interface

Latest News

  • 2026.07.02 Released version 0.8.4 with Doubao TTS API Key setup and legacy credential compatibility
  • 2026.04.03 Released version 0.7.8, refactored the documentary frame-analysis pipeline with a shared service and improved extraction, caching, vision batching, and narration generation
  • 2025.05.11 Released new version 0.6.0, supports short drama commentary and optimized editing process
  • 2025.03.06 Released new version 0.5.2, supports DeepSeek R1 and DeepSeek V3 models for short drama mixing
  • 2024.12.16 Released new version 0.3.9, supports Alibaba Qwen2-VL model for video understanding; supports short drama mixing
  • 2024.11.24 Opened Discord community: https://discord.com/invite/V2pbAqqQNb
  • 2024.11.11 Migrated open source community, welcome to join! Join the official community
  • 2024.11.10 Released official documentation, details refer to Official Documentation
  • 2024.11.10 Released new version v0.3.5; optimized video editing process,

Major Benefits 🎉

From now on, fully support DeepSeek model! Register to enjoy 20 million free tokens (worth 14 yuan platform quota), editing a 10-minute video only costs 0.1 yuan!

🔥 Quick benefits:
1️⃣ Click the link to register: https://cloud.siliconflow.cn/i/pyOKqFCV
2️⃣ Log in with your phone number, be sure to fill in the invitation code: pyOKqFCV
3️⃣ Receive a 14 yuan quota, experience high cost-effective AI editing quickly!

💡 Low cost, high creativity:
Silicon Flow API Key can be integrated with one click, doubling intelligent editing efficiency!
(Note: The invitation code is the only proof for benefit collection, automatically credited after registration)

Immediately take action to unlock your AI productivity with "pyOKqFCV"!

😊 Update Steps: Integration Package: Click update.bat one-click update script Code Build: Use git pull to fetch the latest code

Announcement 📢

Note⚠️: Recently, someone has been impersonating the author on x (Twitter) to issue tokens on the pump.fun platform! This is a scam!!! Do not be deceived! Currently, NarratoAI has not made any official promotions on x (Twitter), please be cautious

Below is a screenshot of this person's x (Twitter) homepage

Screenshot_20250109_114131_Samsung Internet

Future Plans 🥳

  • [x] Windows Integration Pack Release
  • [x] Optimized the story generation process and improved the generation effect
  • [x] Released version 0.3.5 integration package
  • [x] Support Alibaba Qwen2-VL large model for video understanding
  • [x] Support TwelveLabs Pegasus as an optional video-understanding backend (analyzes footage natively to drive highlight selection and commentary; opt-in, set vision_llm_provider = "twelvelabs")
  • [x] Support short drama commentary
  • [x] One-click merge materials
  • [x] One-click transcription
  • [x] One-click clear cache
  • [ ] Support exporting to Jianying drafts
  • [X] Support short drama commentary
  • [ ] Character face matching
  • [ ] Support automatic matching based on voiceover, script, and video materials
  • [ ] Support more TTS engines
  • [ ] ...

System Requirements 📦

  • Recommended minimum: CPU with 4 cores or more, 8GB RAM or more, GPU is not required
  • Windows 10/11 or MacOS 11.0 or above
  • Python 3.12+

Feedback & Suggestions 📢

👏 1. You can submit issue or pull request

💬 2. Join the open source community exchange group

📷 3. Follow the official account [NarratoAI助手] to grasp the latest news

Reference Projects 📚

  • https://github.com/FujiwaraChoki/MoneyPrinter
  • https://github.com/harry0703/MoneyPrinterTurbo

This project was refactored based on the above projects with the addition of video narration features. Thanks to the original authors for their open-source spirit 🥳🥳🥳

Buy the Author a Cup of Coffee ☕️

Image 1 Image 2

License 📝

Click to view LICENSE file

Star History

Star History Chart

Core symbols most depended-on inside this repo

get
called by 1131
app/services/prompts/registry.py
tr
called by 737
webui.py
exists
called by 150
app/services/prompts/manager.py
_summary_state_key
called by 37
webui/components/script_settings.py
_get_subtitle_mask_value
called by 27
webui/components/subtitle_settings.py
remove
called by 23
app/services/prompts/registry.py
register_prompt
called by 17
app/services/prompts/manager.py
_update_video_generation_task
called by 15
app/services/task.py

Shape

Function 788
Method 543
Class 165
Route 2

Languages

Python100%

Modules by API surface

app/services/test_fun_asr_subtitle_unittest.py69 symbols
app/services/jianying_draft_builder.py54 symbols
app/services/fun_asr_subtitle.py54 symbols
app/services/voice.py50 symbols
app/services/generate_video.py50 symbols
webui/components/script_settings.py44 symbols
app/utils/utils.py42 symbols
app/utils/script_generator.py39 symbols
webui/components/audio_settings.py36 symbols
webui/components/subtitle_settings.py35 symbols
app/services/documentary/frame_analysis_service.py33 symbols
app/utils/ffmpeg_utils.py30 symbols

Dependencies from manifests, versioned

Pillow10.3.0 · 1×
azure-cognitiveservices-speech1.37.0 · 1×
dashscope1.24.6 · 1×
edge-tts7.2.7 · 1×
google-generativeai0.8.5 · 1×
loguru0.7.3 · 1×
moviepy2.1.1 · 1×
openai1.77.0 · 1×
pydub0.25.1 · 1×
pysrt1.1.2 · 1×
requests2.32.0 · 1×
streamlit1.56.0 · 1×

For agents

$ claude mcp add NarratoAI \
  -- python -m otcore.mcp_server <graph>

⬇ download graph artifact