Downloads · 30 days
0
Luigi/x-asr-zh-tw-en-streaming-native-demo
x-asr-zh-tw-en-streaming-native-demo is a automatic speech recognition model from Luigi. Use it when you need speech turned into text. It is set up for sherpa-onnx. The card lists the license as apache-2.0.
Streaming zh/en code-switch ASR that outputs native Traditional Chinese (zh-TW) + English directly — no runtime OpenCC step — at the same accuracy as the deployed X-ASR model + OpenCC s2twp.
Downloads · 30 days
0
Access
Public
Updated Jun 21, 2026
Repo size
338 MB
Likes
0
Public
Click a slice to open those files.
.onnx169 MB · 100%
From the Hugging Face model README
Streaming zh/en code-switch ASR that outputs native Traditional Chinese (zh-TW) + English directly —
no runtime OpenCC step — at the same accuracy as the deployed X-ASR model + OpenCC s2twp.
This is the deployed X-ASR 480 ms int8 streaming punct model (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 |
It matches deployed + s2twp (0.0675 vs 0.0683, within bootstrap noise) and is ~2× better than the
earlier native demo, which was built on a weak base checkpoint. Recognition is identical to the deployed model
(orthography-neutralized CER ≈ 0.064); 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.
import sherpa_onnx
rec = sherpa_onnx.OnlineRecognizer.from_transducer(
tokens="tokens.txt", encoder="encoder.int8.onnx",
decoder="decoder.onnx", joiner="joiner.int8.onnx",
num_threads=2, provider="cpu", decoding_method="greedy_search")
# feed 16 kHz mono audio; pad ~2 s trailing silence to flush the streaming chunk.
# output is Traditional zh-TW + English, no OpenCC needed.
s2twp on everyday speech.