OpenFormosa台灣取向模型訓練與評測工作站，先做 OpenFormosa-Base，再接 ASR、TTS、OCR 和台灣在地 benchmark。https://openformosa.com/2026-06-23T00:00:00+08:00BlueMagpie-TTS: Taiwanese-accent, Chinese–English code-switching speech synthesishttps://openformosa.com/blog/2026/06/23/bluemagpie-tts/2026-06-23T00:00:00+08:002026-06-23T00:00:00+08:00An open Taiwanese-accent text-to-speech model that handles Chinese–English code-switching — keep VoxCPM's acoustic stack, swap in the Barbet language model, and cut character error rate by about 58% on a hard test set.Barbet 1B Base: a hybrid decoder-only language model for Traditional Chinesehttps://openformosa.com/blog/2026/06/21/barbet-1b-base/2026-06-21T00:00:00+08:002026-06-21T00:00:00+08:00A 1B-parameter hybrid decoder-only causal language model — global and sliding-window attention interleaved with Mamba, context up to 1M, embedding tying, built on PangolinTokenizer.PangolinTokenizer: a byte-level BPE tokenizer for Traditional Chinese and Taiwanhttps://openformosa.com/blog/2026/06/20/pangolin-tokenizer/2026-06-20T00:00:00+08:002026-06-20T00:00:00+08:00A byte-level BPE tokenizer built for Taiwan — 114,688 merges, the lowest tokens/character on PangolinBench with the smallest vocabulary.