---
license: apache-2.0
base_model: FireRedTeam/FireRedASR2-AED
pipeline_tag: automatic-speech-recognition
library_name: openasr
tags:
- automatic-speech-recognition
- speech-to-text
- openasr
- oasr
- firered-aed-l-v2
---
# FireRedASR2 AED-L Β· OpenASR
**FireRedTeam's Mandarin-first bilingual ASR β attention encoder-decoder tuned for state-of-the-art Chinese and dialect accuracy**
[](https://huggingface.co/FireRedTeam/FireRedASR2-AED/blob/main/README.md)
[](https://github.com/QuintinShaw/openasr)
[](https://openasr.org)
[](https://huggingface.co/FireRedTeam/FireRedASR2-AED)
Native speech-to-text in the **[OpenASR](https://github.com/QuintinShaw/openasr)** runtime β
engineered for peak performance on CPU & GPU, **no Python at inference time**.
---
## β¨ Highlights
- π₯ **State-of-the-art Mandarin accuracy** β 3.05% average CER across four public Mandarin benchmarks, with 0.57% CER on aishell1 (arXiv:2603.10420 Table 2)
- π£οΈ **Strong dialect coverage** β 11.67% average CER across 19 public Chinese dialect/accent benchmarks, beating Doubao-ASR (15.39%) and matching Qwen3-ASR (11.85%) on the paper's own comparison table
- π¨π³π¬π§ **Bilingual Mandarin + English** β one 1.1B checkpoint decodes both languages plus code-switching, no language flag needed
- π€ **Singing-lyrics robustness** β the paper reports 1.17% CER on a singing-lyrics benchmark, alongside the speech-recognition numbers above
- ποΈ **Conformer encoder + Transformer decoder** β an attention-based encoder-decoder, architecturally identical to FireRedASR-AED-L, retrained on a larger vocabulary
- π¦ **Native in OpenASR** β `.oasr` packs run with no Python at inference, engineered for peak performance on CPU & GPU
## π Quickstart
```bash
# 1. Install the OpenASR CLI Β· https://openasr.org
# 2. Pull a build (pick a quant β see the table below)
openasr pull firered-aed-l-v2:q4
# 3. Transcribe
openasr transcribe audio.wav --model firered-aed-l-v2
```
All builds for this model:
```bash
openasr pull firered-aed-l-v2:fp16
openasr pull firered-aed-l-v2:q8
openasr pull firered-aed-l-v2:q4
```
## π¦ Available builds
| Quant | File (`.oasr`) | Size | RAM peak | RTF Β· M1 CPU | RTF Β· M1 GPU | JFK ΞWER vs fp16 |
|:------|:---------------|-----:|---------:|-------------:|-------------:|-----------------:|
| fp16 | `firered-aed-l-v2-fp16.oasr` | 2.35 GB | 4.74 GB | 0.40Γ | 0.79Γ | 0.0% |
| q8_0 | `firered-aed-l-v2-q8_0.oasr` | 1.28 GB | 4.35 GB | 0.36Γ | 0.35Γ | 0.0% |
| q4_k | `firered-aed-l-v2-q4_k.oasr` | 709 MB | 4.10 GB | 0.38Γ | 0.31Γ | 0.0% |
RTF = real-time factor on the fixed 11s JFK clip (**lower is faster**); RAM peak measured per pack
in an isolated subprocess. JFK ΞWER compares each quantized build's JFK transcript to this model's
fp16 JFK transcript, so it measures quantization drift rather than absolute recognition accuracy.
**q4_k** is the recommended default β near-reference quality at a fraction of the
footprint.
## π§ About FireRedASR2 AED-L
FireRedASR2-AED is the attention-based encoder-decoder member of **FireRedASR2**, the successor
to FireRedTeam's open-source industrial-grade **FireRedASR** speech-recognition family, released
as part of the **FireRedASR2S** all-in-one ASR system. It pairs a **Conformer encoder** with a
**Transformer decoder** at 1.1B parameters -- architecturally byte-identical to the original
FireRedASR-AED-L -- retrained on a larger token vocabulary. The FireRedASR2S technical report
(arXiv:2603.10420, Table 2) reports **3.05% average Character Error Rate** across four public
Mandarin benchmarks (aishell1 at 0.57%), **11.67% average CER** across 19 public Chinese
dialect/accent benchmarks, and **1.17% CER** on a singing-lyrics benchmark, outperforming
Doubao-ASR and matching Qwen3-ASR on the paper's own comparison table. It is bilingual (Mandarin
Chinese and English, including code-switching). 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 **q4_k** build is the recommended default for everyday use; **q8_0** trades
some size for closer fp16 fidelity and **fp16** is for maximum fidelity or verification. Note:
FireRedTeam also publishes a larger **FireRedASR2-LLM** (Encoder-Adapter-LLM, 2.89%/11.55% avg
CER) variant that this repo does not distribute -- the numbers above are for this AED checkpoint
only, and should not be conflated with the original (v1) FireRedASR-AED-L's 3.18%/0.55% figures,
which come from a different training run and benchmark protocol.
## βοΈ How these packs were made
Converted from [FireRedTeam/FireRedASR2-AED](https://huggingface.co/FireRedTeam/FireRedASR2-AED) with the OpenASR importer:
```bash
openasr model-pack import firered-aed .oasr \
--package-id firered-aed-l-v2 --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.
## βοΈ License
These packs **inherit the upstream model's license: Apache-2.0**
([source](https://huggingface.co/FireRedTeam/FireRedASR2-AED/blob/main/README.md)). OpenASR packaging retains the upstream copyright and
NOTICE; the only modifications are format conversion and quantization.
## π Acknowledgements
This pack is a redistribution of **FireRedASR2-AED**, created and released by **FireRedTeam**
([FireRedTeam/FireRedASR2-AED](https://huggingface.co/FireRedTeam/FireRedASR2-AED),
[FireRedTeam/FireRedASR2S](https://github.com/FireRedTeam/FireRedASR2S)). All credit for the
architecture, training, and weights belongs to FireRedTeam; the license is inherited from and
identical to the upstream model (**Apache-2.0**, as declared on the upstream model card). Thank
you to FireRedTeam for releasing their work openly. OpenASR only performs format conversion,
quantization, runtime verification, and local-inference adaptation.
## π Links
- π¦ **OpenASR** β
- π **Website** β
- π€ **Upstream model** β [FireRedTeam/FireRedASR2-AED](https://huggingface.co/FireRedTeam/FireRedASR2-AED)