slvDev
Featured
esp32-ai
A 28.9 million parameter language model runs on an $8 ESP32-S3 microcontroller entirely on-device, generating simple stories at about 9.5 tokens per second.
AI repository intelligence
FR-AI / ANALYSISWhy it is worth attention
It adapts Meshtastic to Traditional Chinese with zhuyin input on select devices, backed by checked-in generated data and clear license separation.
Who it is for
- Traditional Chinese-speaking Meshtastic users
- Owners of the four listed boards (GAT562, Seeed Wio Tracker L1, Heltec V3, Cardputer Adv)
- Bopomofo / zhuyin input users
- Meshtastic firmware tinkerers seeking unofficial localized builds
Use cases
- Communicating in Traditional Chinese or Bopomofo over LoRa mesh on supported boards
- Using zhuyin on-screen or physical keyboard input for messaging on Cardputer, GAT562, or Wio Tracker
- Displaying Traditional Chinese text on Heltec LoRa32 V3
- Building a customized zhtw firmware environment locally with pio run
Strengths
- Adds Bopomofo input on three board variants and Traditional Chinese rendering on all four
- Generated glyph and dictionary data is checked in, so firmware builds need no extra tooling
- Does not redistribute font files, only embedded bitmaps, and includes license details
- Keeps all stock upstream environments untouched, limiting accidental behavior changes
Considerations
- Unofficial and not endorsed by the Meshtastic project; no warranty is provided
- Supports only four environments, and Heltec V3 is display-only with no input
- Based on a single upstream release v2.7.26.54e0d8d
README quick start
Traditional Chinese firmware (unofficial)
This is an unofficial derivative of meshtastic/firmware,
based on release v2.7.26.54e0d8d. It adds Traditional Chinese text rendering and a
Bopomofo (zhuyin) input method to four boards, and nothing else: every stock
environment is left exactly as upstream ships it.
It is not affiliated with or endorsed by the Meshtastic project. Meshtastic® is a registered trademark of Meshtastic LLC. Please report problems with these builds here rather than to the upstream project, and reproduce them on a stock build before reporting anything upstream.
| Environment | Board | Input |
|---|---|---|
gat562_family_zhtw | GAT562 Family (nRF52840) | Bopomofo on the on-screen keyboard |
seeed_wio_tracker_L1_zhtw | Seeed Wio Tracker L1 / L1 Lite / L1 Pro | Bopomofo on the on-screen keyboard |
heltec-v3_zhtw | Heltec LoRa32 V3 | display only |
m5stack-cardputer-adv_zhtw | M5Stack Cardputer Adv | Bopomofo on the physical keyboard |
pio run -e gat562_family_zhtw
The glyph tables and the dictionary are generated data, checked in so the firmware
builds without extra tooling. No font file is redistributed; only rasterised bitmaps
are embedded. Terms for every font and dictionary in the chain are in
licenses/, alongside the upstream GPL-3.0 licence that covers the
firmware as a whole.
These builds come with no warranty of any kind, express or implied. Flashing firmware can leave a device unusable, and a mesh radio running unverified firmware may behave in ways you do not expect. Use them at your own risk.
繁體中文韌體（非官方）
這是 meshtastic/firmware v2.7.26.54e0d8d
的非官方衍生版本，替四款機種加上繁體中文顯示與注音輸入法，其餘一律不動：所有原廠
env 都與上游完全相同。
本專案與 Meshtastic 官方專案無隸屬關係，亦未經其背書。Meshtastic® 為 Meshtastic LLC 的註冊商標。這些韌體如有問題請回報到本專案，不要送去官方；要向官方回報之前，請先在 原廠韌體上重現。
字型點陣表與注音字典都是產生出來的資料，直接收進版本庫，所以編譯時不需要額外工具。
本專案不轉散布任何字型檔，嵌入的只有點陣資料；所有字型與字典的授權條款放在
licenses/，韌體整體則沿用上游的 GPL-3.0。
本韌體不提供任何形式的保證。刷機有機會讓裝置無法開機，跑著未經驗證韌體的無線電 節點也可能出現非預期行為。請自行承擔使用風險。
以下為上游 Meshtastic 專案的原始 README，其中的徽章與連結皆指向上游專案。
Meshtastic Firmware
Description
Unofficial Meshtastic firmware with Traditional Chinese display and a Bopomofo input method
Related repositories
Similar projects matched by category, topics, and programming language.
m-novotny
Featured
memguard-rs
A Rust library that provides secure memory handling primitives including zeroization on drop, memory locking, constant-time comparison, and compile-time guarded regions, with zero dependencies and no_std support.
zju3dv
Featured
INTACT-JEPA
INTACT is an end-to-end JEPA world model that learns an isomorphic intent-to-action interface, enabling search-free direct control for goal-conditioned robot tasks with high success rates after just one epoch of training.