Views
No views yet
s2twp.encoder.int8.onnx / decoder.onnx /
joiner.int8.onnx, unchanged) with its tokens.txt relabeled: each Chinese token surface is mapped
Simplified→Taiwan-Traditional with OpenCC s2twp. The model emits the same token IDs as the deployed model;
the relabeled tokenizer renders them as Traditional. The OpenCC conversion is baked into the tokenizer, so
there is zero post-processing at inference time and zero speed cost (it is the deployed model).| Pipeline | Traditional CER | Runtime OpenCC |
|---|---|---|
deployed X-ASR + OpenCC s2twp | 0.0683 | yes (post-step) |
| this model (native Traditional) | 0.0675 | none |
| prior native demo (weak base, relabel) | 0.137 | none |
s2twp adds essentially no orthography error, so a native model cannot
do better — this reaches that ceiling while removing the post-step. Real-time on a Jetson Nano at 2 CPU threads.1import sherpa_onnx
2rec = sherpa_onnx.OnlineRecognizer.from_transducer(
3 tokens="tokens.txt", encoder="encoder.int8.onnx",
4 decoder="decoder.onnx", joiner="joiner.int8.onnx",
5 num_threads=2, provider="cpu", decoding_method="greedy_search")
6# feed 16 kHz mono audio; pad ~2 s trailing silence to flush the streaming chunk.
7# output is Traditional zh-TW + English, no OpenCC needed.s2twp on everyday speech.