llll-view

An asset viewer for the LLLL game. Research purpose only.

I cannot take responsibility. Please be responsible for yourself. Do not host publicly as a service.

This project is still in development. Contributions are very welcome. Linux is supported.

Features

Audio - ◎

• convert from acb to m4a
• playable with proper metadata
• get new songs when released

Card - ◎

• show illustration
• show card effect movie
• play card voice
• show card status
• translation support
  • ongoing, partly supported

Story - ○

• show storyline
• play voice
• search any word from storyline
• translation support
• show 3d model
• move camera

Card game - ×

• show card game assets
• playable card game simulator

Rhythm game - ×

• show rhythm game assets
• playable rhythm game simulator

With×Meets - ×

• show with×meets assets
• playable with×meets simulator
• translation support

Fes×Live - ×

• show fes×live assets
• playable fes×live simulator

Setup

If you are new to web development, please follow the steps below.

Your final directory layout should look like this:

llll/
  llll-tools/
    bin/
      ffmpeg
    vgmstream/
      build/
        cli/
          vgmstream-cli
    AssetStudio/
      AssetStudioCLI/
        bin/
          Release/
            net8.0/
              AssetStudioModCLI
    UsmToolkit/
      UsmToolkit
  llll-view/
    ...
  inspix-hailstorm/
    ...

After cloning, check backend/.env to understand required tool names and paths.

Install Node.js and pnpm

I recommend using proto to manage Node.js and pnpm versions.

# Ubuntu / Debian
apt-get install git unzip gzip xz-utils

And then, install proto:

bash <(curl -fsSL https://moonrepo.dev/install/proto.sh)

Setup project dependencies

This project requires many dependencies. First, decide the root project directory (for example, ~/llll) and clone this repository into it:

git clone (this repository url)

Install llll-view dependencies

Install node dependencies using pnpm:

cd llll-view
pnpm install

Install llll-tools

The tools below are required to process LLLL assets. Place them under llll/llll-tools.

• ffmpeg (to convert various audio formats)
• vgmstream (to convert from original acb format to wav)
• AssetStudioCLI (this version is modified to support recent Unity versions)
• UsmToolkit (to extract card effect movies)

Some tools depend on:

• .NET 3.1 (required by UsmToolkit)
• .NET 8.0 (required by AssetStudioCLI)

Please be careful about versions of .NET.

You can get these tools as open source projects.

However, due to licensing issues, official prebuilt binaries do not include the good m4a library. If you want high-quality m4a output, build ffmpeg yourself and place it at llll-tools/bin/ffmpeg. If you do not mind m4a quality, use an official static build instead.

WebP support

Some thumbnail images are converted to WebP format, so cwebp command is required:

# Ubuntu/Debian
sudo apt-get install webp

Frontend setup

Copy frontend/.env.example to frontend/.env and adjust variables if necessary. CORS_ORIGIN refers to backend server origin. This is required to allow browser to access backend server.

Backend setup

Copy backend/.env.example to backend/.env and adjust variables if necessary.

Asset server setup

Thanks to inspix-hailstorm, you can handle assets like official client. Clone the repository under llll/ directory:

cd llll
git clone https://github.com/vertesan/inspix-hailstorm

Final check

If you are not sure whether setup is correct, check backend/.env for required tool names and paths, then adjust them to your environment.

Browser requirements

Google Chrome on PC is required for some features like transcription and translation. Official requirements are here: https://developer.chrome.com/docs/ai/prompt-api

• Translation

Set chrome://flags/#translation-api to Enable and use an HTTPS connection to use the translation feature.

• Transcription

Set the following flags to Enable: chrome://flags/#optimization-guide-on-device-model chrome://flags/#prompt-api-for-gemini-nano chrome://flags/#prompt-api-for-gemini-nano-multimodal-input

I tested with an RTX 5070 Ti. Speed is pretty good, but quality is quite low. Whisper might be better, but it adds heavy dependencies on the server or client side.

FAQ

Please check help page in the app first.

If you run into issues during setup, use an AI chat to troubleshoot.

I do not have any rights to LLLL assets.