README.md

---
license: apache-2.0
base_model: GilgameshWind/X-ASR-zh-en
pipeline_tag: automatic-speech-recognition
library_name: openasr
tags:
  - automatic-speech-recognition
  - speech-to-text
  - openasr
  - oasr
  - xasr
---

<div align="center">

# X-ASR zh-en · OpenASR

**Bilingual Chinese + English streaming speech recognition — a compact icefall Zipformer2 transducer**

[![License](https://img.shields.io/badge/license-Apache--2.0-2563eb.svg)](https://huggingface.co/GilgameshWind/X-ASR-zh-en/blob/main/README.md)
[![Format](https://img.shields.io/badge/format-.oasr-7c3aed.svg)](https://github.com/QuintinShaw/openasr)
[![Runtime](https://img.shields.io/badge/runtime-OpenASR-111827.svg)](https://openasr.org)
[![Base model](https://img.shields.io/badge/base-X--ASR--zh--en-f59e0b.svg)](https://huggingface.co/GilgameshWind/X-ASR-zh-en)

Native speech-to-text in the **[OpenASR](https://github.com/QuintinShaw/openasr)** runtime —
engineered for peak performance on CPU & GPU, **no Python at inference time**.

</div>

---

## ✨ Highlights

- 🇨🇳🇬🇧 **Chinese + English** — one bilingual checkpoint for zh/en speech, including code-switched audio
- ⚡ **Streaming-first, offline-capable** — a cache-aware streaming Zipformer2 transducer for low-latency captions that also runs full-file offline transcription
- 🪶 **Compact ~0.16B** — a 6-stack Zipformer2 encoder + stateless RNN-T decoder + tanh joiner over a 5000-token BPE vocab, light enough for on-device CPU
- 🦀 **Native in OpenASR** — `.oasr` packs run with no Python at inference, engineered for peak performance on CPU & GPU

## 🚀 Quickstart

```bash
# 1. Install the OpenASR CLI  ·  https://openasr.org
# 2. Pull a build (pick a quant — see the table below)
openasr pull xasr-zh-en:q8

# 3. Transcribe
openasr transcribe audio.wav --model xasr-zh-en
```

All builds for this model:

```bash
openasr pull xasr-zh-en:fp16
openasr pull xasr-zh-en:q8
openasr pull xasr-zh-en:q4
```

## 📦 Available builds

| Quant | File (`.oasr`) | Size | RAM peak | RTF · M1 CPU | RTF · M1 GPU | JFK ΔWER vs fp16 |
|:------|:---------------|-----:|---------:|-------------:|-------------:|-----------------:|
| fp16 | `xasr-zh-en-fp16.oasr` | 315 MB | 824 MB | 0.07× | 0.12× | 0.0% |
| q8_0 | `xasr-zh-en-q8_0.oasr` | 176 MB | 549 MB | 0.06× | 0.12× | 0.0% |
| q4_k | `xasr-zh-en-q4_k.oasr` | 112 MB | 423 MB | 0.06× | 0.12× | 0.0% |

<sub>RTF = real-time factor on the fixed 11s JFK clip (**lower is faster**); RAM peak measured per pack
in an isolated subprocess. JFK ΔWER compares each quantized build's JFK transcript to this model's
fp16 JFK transcript, so it measures quantization drift rather than absolute recognition accuracy.
**q8_0** is the recommended default — near-reference quality at a fraction of the
footprint.</sub>

## 🧠 About X-ASR zh-en

X-ASR-zh-en is a compact **bilingual (Chinese + English)** streaming speech-recognition model from
**GilgameshWind**, built with the **icefall / k2** recipe as a cache-aware **Zipformer2 RNN-T
transducer** (a 6-stack, 19-layer Zipformer2 encoder, a stateless RNN-T decoder, and a tanh joiner
over a 5000-token BPE vocabulary, ~0.16B parameters). The same checkpoint serves both low-latency
**streaming** captions and full-file **offline** transcription, making it a good fit for on-device
Chinese/English dictation and real-time subtitles. This OpenASR repo repackages the weights as
`.oasr` packs that run natively in the OpenASR runtime — no Python at inference time, all decoding
local. The **q8_0** build is the recommended default (it matched the fp16 transcript bit-for-bit in
OpenASR's verification); **q4_k** is the smallest build for tight-memory devices and **fp16** is for
maximum fidelity or verification.

## ⚙️ How these packs were made

Converted from [GilgameshWind/X-ASR-zh-en](https://huggingface.co/GilgameshWind/X-ASR-zh-en) with the OpenASR importer:

```bash
openasr model-pack import xasr-zipformer <src> <out>.oasr \
  --package-id xasr-zh-en --quantization {fp16,q8-0,q4-k}
```

The `.oasr` container is GGUF-backed; packs use zero-copy mmap weight binding and graph
buffer reuse to keep peak memory low.

## ⚖️ License

These packs **inherit the upstream model's license: Apache-2.0**
([source](https://huggingface.co/GilgameshWind/X-ASR-zh-en/blob/main/README.md)). OpenASR packaging retains the upstream copyright and
NOTICE; the only modifications are format conversion and quantization.

## 🙏 Acknowledgements

This pack is a redistribution of **X-ASR-zh-en**, created and open-sourced by **GilgameshWind**
([GilgameshWind/X-ASR-zh-en](https://huggingface.co/GilgameshWind/X-ASR-zh-en)). All credit for the
original architecture, training, and weights belongs to the author; the license is inherited from and
identical to the upstream model (Apache-2.0). The model is built on the **icefall** /
**k2 / Next-gen Kaldi** Zipformer2 transducer recipe — thank you to the icefall team and to
GilgameshWind for releasing their work openly. OpenASR only performs format conversion, quantization,
runtime verification, and local-inference adaptation.

## 🔗 Links

- 🦀 **OpenASR** — <https://github.com/QuintinShaw/openasr>
- 🌐 **Website** — <https://openasr.org>
- 🤗 **Upstream model** — [GilgameshWind/X-ASR-zh-en](https://huggingface.co/GilgameshWind/X-ASR-zh-en)