kodama-ja-streaming-small
A Japanese streaming ASR model adapted from the English streaming model moonshine-streaming-small.
I build Japanese speech AI and the characters and games that use it — on-device Japanese TTS, voice dialogue, and Unity integration.
Japanese TTS, G2P and pitch accent, speech-to-speech dialogue, and on-device inference — from model training to Unity / C# integration.
A Japanese streaming ASR model adapted from the English streaming model moonshine-streaming-small.
piper-plus, a lightweight Japanese-capable TTS, and uPiper, a library for using it from Unity.
A clean pretrained model for Style-Bert-VITS2, published on Hugging Face.
AI characters that combine voice dialogue with character expression, such as desktop mascots and conversational apps.
An open-source, user-customizable desktop mascot.
An AI agent character that picks up YouTube comments and keeps the conversation going.
Game development in Unity and games and prototypes with generative AI built in. Organizer of a generative AI exhibition.
An exploration adventure set in a hidden magical academy, where I joined as PM / director.
I organize the Generative AI Nandemo Exhibition, a demo event for individual creators building with generative AI, and handle its planning and operations.
An ADV game set in a flower shop where you make bouquets (July to November 2023).
Libraries for Unity, development infrastructure, and community work.
Running MidraLab, a community for proofs of concept and technical validation shared publicly.
A library that makes Google Sheets master data easy to use in Unity.
A library that makes building ADV games in Unity easier with Fungus and Google Sheets.
As of 2026-09-29