ACUAH：ユーザーカスタマイズ可能な3Dキャラアシスタント

CsUnite Corporation

About this app

Application instructions are posted on the ACUAH homepage.
　https://acuah.info
　https://riemgoshawk.booth.pm/

[Caution]
If you wish to use ACUAH commercially, a commercial use license is required. Please contact us.

Create a 3D character assistant with your “likes”
We will continue to add character motion data and functions in the future.

Character model: VRM0.0/VRM1.0 format
・Standard model (ACUAH)
・VRM file reading
・VRoidHub (3D character platform of Pixiv Co., Ltd.) cooperation

Dialogue scenario/function: CSV format
・Standard dialogue scenarios are automatically downloaded when the app starts.
The main functions are as follows.
- Clock, timer
- Weather forecast (Japan only)
- Notepad
- Music playback (mp3/wav/ogg)
- Dance(Unite in the sky (C)UTJ/UCL)
- Photo shoot/photo album
- QR code reading
- Making a phone call
- Photography posing
- IFTTT Webhooks collaboration/ROS2 collaboration
- Rock, paper, scissors
・LLM/Voice API (VOICEVOX, Microsoft Azure) cooperation
・LLM Function calling compatible

Audio data: MP3 format
・The following audio data is available.
- Nana Shiina (twitter @C7na7)
- Mr. Tsukisha Rei (twitter @keiiiiin_orz)
- Ah-ya (twitter @aya_voicer) (English)

Misskey cooperation (test implementation)
・Characters can post and reply.
*We recommend creating and setting up a character-specific account.

Pseudo hologram
・If you combine it with a smartphone stand that uses the Pepper's Ghost mechanism, you can experience a little bit of autostereoscopic vision.

I want to freely replace dialogue scenarios, audio data, etc.!
With ACUAH editor, you can replace dialogue scenarios, audio data, etc. with the original ones.

ACUAH editor
・Windows (x64) version, macOS version Dialogue scenario, voice data, etc. replacement data creation tool for ACUAH
BOOTH https://riemgoshawk.booth.pm/

----------------------------------------------------------------

Character models, voice data
Create a 3D character assistant with your "likes"!

Application manual can be found on the website.
Please read the manual before use.
ACUAH: https://acuah.info
BOOTH:　https://riemgoshawk.booth.pm/

CAUTION
If you want to use "ACUAH" for commercial use, you need to purchase a commercial use license. Please ask.

Character model: VRM0.0/VRM1.0
-Default VRM model (ACUAH)
- Any VRM files can be loaded.
- It works with VRoidHub (pixiv Inc.), the 3D character platforms.
You can also create your favorite characters with VRoid Studio, VRoid Mobile (pivix Inc.).

Dialogue scenario, functions: CSV
The standard scenario will be downloaded automatically when the application is launched.
The main features supported by latest standard scenarios are as follows
- Clock, timer
- Notepad
- Music playback (mp3/wav/ogg)
- Dance (Unite in the sky (C)UTJ/UCL)
- Photo and Photo album
-QR code reader
- Phone dial-out
- Posing for screenshots
- IFTTT Webhooks linkage/ROS2 linkage
- Play rock-paper-scissors

LLM/Microsoft Azure Cognitive service (Text-To-Speech) support. (require your API key)
LLM Function calling support.

Voice data files: MP3
The following voice data is available for use with the standard scenario.
- Nana Shiina (Twitter @C7na7)
- Rei Tsukisha (Twitter @keiiiiin_orz)
- Ah-ya (Twitter @aya_voicer) (English)

Misskey integration
*Test implementation
The character sometimes posts or replies to Following.
We recommend that you create and set up a dedicated account for your character.

False-Hologram
If you combine a smartphone stand that uses the Pepper's Ghost, you can experience a little bit of stereoscopic vision.

Freely replace dialogue scenario and voice data!
ACUAH editor
Windows (x64) and macOS applications.
With "ACUAH editor" on BOOTH, you can replace the dialogue scenario and voice data with the original ones.

Please feel free to create your own assistant.
More character motion/voice data and features will be added in the future.

What’s new

ver.3.8.0
* Added user registration functionality (name, face recognition)
* Unified character voices to synthetic speech when using the "Scenario/LLM" dialog method
* Renamed the "Scenario/LLM" dialog method to "LLM function calling"
* Bug fixes
- Minor bug fixes
- Voice Settings > VOICEOVOX: test audio did not play when using an external web API
- AI (LLM) Llama does not work properly