What is this?
NarratoAI is an all-in-one AI tool for film/show narration and automated video editing: it uses an LLM to write the narration script, then automatically edits the footage, generates voice-over, and adds subtitles.
NarratoAI is an all-in-one AI tool for film/show narration and automated video editing: it uses an LLM to write the narration script, then automatically edits the footage, generates voice-over, and adds subtitles.Is this a desktop app or something I access through a browser?It ships as a downloadable Windows/macOS integration package that launches a local server, and you use it through a web UI in your browser; there's also a separately hosted cloud version at NarratoAI.co if you don't want to run it yourself.
It ships as a downloadable Windows/macOS integration package that launches a local server, and you use it through a web UI in your browser; there's also a separately hosted cloud version at NarratoAI.co if you don't want to run it yourself.How do I get it running?Download the integration package for your OS (NarratoAI-windows-x64 or NarratoAI-macos-arm64), unzip it fully without moving the internal folders, run the update script (update-windows.bat / update-macos.command), then the start script, and open the web UI in your browser. A macOS Docker deployment and a plain local-run option are also available.
Download the integration package for your OS (NarratoAI-windows-x64 or NarratoAI-macos-arm64), unzip it fully without moving the internal folders, run the update script (update-windows.bat / update-macos.command), then the start script, and open the web UI in your browser. A macOS Docker deployment and a plain local-run option are also available.What do I need before running it?You need Python 3.12+, Windows 10/11 or macOS 11.0+, at least a 4-core CPU and 8GB RAM; a GPU is not required. You'll also need API keys for whichever LLM/TTS providers you configure (e.g. DeepSeek, Gemini, Qwen2-VL).
You need Python 3.12+, Windows 10/11 or macOS 11.0+, at least a 4-core CPU and 8GB RAM; a GPU is not required. You'll also need API keys for whichever LLM/TTS providers you configure (e.g. DeepSeek, Gemini, Qwen2-VL).Any current gaps I should know about?The Windows build of the IndexTTS-2 MLX voice-cloning pack isn't released yet (only the macOS version is), and the project recently dropped its LiteLLM dependency in favor of OpenAI-compatible requests only, for security reasons.
The Windows build of the IndexTTS-2 MLX voice-cloning pack isn't released yet (only the macOS version is), and the project recently dropped its LiteLLM dependency in favor of OpenAI-compatible requests only, for security reasons.Why does the README warn about scams?People have reportedly renamed this free, open-source project and resold it on social platforms like Douyin and Bilibili, so the authors explicitly warn users not to pay for it.
People have reportedly renamed this free, open-source project and resold it on social platforms like Douyin and Bilibili, so the authors explicitly warn users not to pay for it.Which LLMs can generate the narration script?It supports DeepSeek R1/V3, Alibaba's Qwen2-VL for video understanding, Gemini, and other OpenAI-compatible providers; TwelveLabs Pegasus can optionally be used as the video-understanding backend.
It supports DeepSeek R1/V3, Alibaba's Qwen2-VL for video understanding, Gemini, and other OpenAI-compatible providers; TwelveLabs Pegasus can optionally be used as the video-understanding backend.Does it support voice cloning?Yes, through IndexTTS-1.5, an MLX-accelerated IndexTTS-2 build for Apple Silicon, and the OmniVoice engine, alongside regular TTS options like Tencent Cloud and Doubao.
Yes, through IndexTTS-1.5, an MLX-accelerated IndexTTS-2 build for Apple Silicon, and the OmniVoice engine, alongside regular TTS options like Tencent Cloud and Doubao.What else can it do besides full-length narration?It can also auto-mix short-drama clips, one-click transcribe subtitles, merge source clips, export Jianying (CapCut) drafts, and auto-clean its cache.
It can also auto-mix short-drama clips, one-click transcribe subtitles, merge source clips, export Jianying (CapCut) drafts, and auto-clean its cache.What are similar projects?MoneyPrinterTurbo, kirara-ai, Pixelle-Video, pyvideotrans, AstrBot, Easel, WanGP, GPT-SoVITS
MoneyPrinterTurbo, kirara-ai, Pixelle-Video, pyvideotrans, AstrBot, Easel, WanGP, GPT-SoVITSWhich topics is it in?gemini-api, voice-cloning, video-editing, webui, discord, text-to-speech, qwen, video-generation, apple-silicon, deepseek, gemini, chinese
gemini-api, voice-cloning, video-editing, webui, discord, text-to-speech, qwen, video-generation, apple-silicon, deepseek, gemini, chinese