stack-chan
Here are 15 public repositories matching this topic...
559 K パラメータの日本語 TTS を ESP32-S3 で実時間合成。漢字かな交じり文の形態素解析・アクセント推定まで端末内で走る(M5Stack CoreS3 実機で確認)。推論は依存ゼロの C99、ブラウザ demo あり。arXiv:2608.21378 sanoTTS の日本語 clean-room 再実装。
-
Updated
Oct 1, 2026 - Python
Custom firmware and a self-hosted voice server for the M5Stack StackChan (CoreS3 with two Feetech SCS0009 servos). It replaces the stock firmware.
-
Updated
Oct 9, 2026 - Dockerfile
한국판 스택짱 — M5Stack CoreS3 한국어 가족 AI 로봇 펌웨어 (RoboEyes 눈, 한국어 음성대화, 외부 데이터, 머리 모션 레코더). AI_StackChan_Ex/Stack-chan 포크.
-
Updated
Jun 1, 2026 - C++
Local LLM + voice-cloning TTS sharing one 8 GB GPU, as a server for ESP32 voice boards. 本地 LLM + 声音克隆 TTS 共用一块 8 GB 显卡,给 ESP32 语音板子当服务端。怎么省显存、怎么轮流、怎么不断音,附完整实测。
-
Updated
Oct 17, 2026 - Python
Shows what your coding agent is doing, on a Stack-chan with an M5Stack CoreS3. Works with Codex and Claude Code.
-
Updated
Aug 6, 2026 - C++
M5Stack CoreS3 で日本語ニューラル TTS sanoTTS-jp をローカル実行する Stack-chan ファームウェア (ESP-IDF)。ストリーミング合成 + アバター リップシンク + 端末内漢字 G2P。ciniml/stackchan-idf のフォーク
-
Updated
Sep 7, 2026 - C++
ESP-IDF firmware for a Stack-chan on the M5Stack AtomS3R. Every decision lives in a layer with no ESP-IDF in it, so the whole decision layer is tested on a PC without a board; speech and language models live behind a documented gateway contract.
-
Updated
Oct 17, 2026 - C++
Tiny Japanese language models that run entirely on an ESP32-S3. Action LM (3.15M params, INT4) turns Japanese requests into robot actions on M5Stack Stack-chan.
-
Updated
Oct 4, 2026 - Python
Unofficial browser color simulator for the 3D-printed M5Stack Stack-chan (K151 / CoreS3). AS-IS, no support.
-
Updated
Jul 7, 2026 - HTML
.NET SDK and reference host for the AtomS3R stack-chan voice conversation gateway
-
Updated
Oct 18, 2026 - C#
Add this topic to your repo
To associate your repository with the stack-chan topic, visit your repo's landing page and select "manage topics."