Upload README.md with huggingface_hub

README.md

@@ -0,0 +1,98 @@
+---
+license: cc-by-nc-4.0
+language:
+- sna
+base_model: facebook/omniASR-LLM-300M
+tags:
+- automatic-speech-recognition
+- african-languages
+- waxal
+- omnilingual-asr
+- fairseq2
+datasets:
+- google/WaxalNLP
+metrics:
+- wer
+- cer
+---
+
+# omniasr-llm-300m-waxal-sna
+
+Fine-tune of Meta's **Omnilingual ASR** (wav2vec2_llama (autoregressive LLM decoder)) on **Shona** conversational
+speech from the [WAXAL corpus](https://huggingface.co/datasets/google/WaxalNLP),
+part of a 19-language WAXAL×Omnilingual ASR benchmark.
+
+## Results (held-out WAXAL test set)
+
+| Metric | Score |
+|---|---|
+| WER | 21.6% |
+| UER (unit/char error rate) | 3.7% |
+
+Fine-tuned for 5,000 steps from [`facebook/omniASR-LLM-300M`](https://huggingface.co/facebook/omniASR-LLM-300M) on 2×H200.
+
+## Usage
+
+```python
+# pip install git+https://github.com/facebookresearch/omnilingual-asr.git
+from pathlib import Path
+import torch
+from huggingface_hub import snapshot_download
+from fairseq2.data.tokenizers.hub import load_tokenizer
+from omnilingual_asr.models.inference.pipeline import ASRInferencePipeline
+from omnilingual_asr.models.wav2vec2_llama.hub import get_wav2vec2_llama_model_hub as get_hub
+
+ckpt = snapshot_download("waxal-benchmarking/omniasr-llm-300m-waxal-sna")
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+dtype = torch.bfloat16 if device.type == "cuda" else torch.float32
+
+hub = get_hub()
+config = hub.get_arch_config("300m")
+model = hub.load_custom_model(Path(ckpt) / "model", config, device=device, dtype=dtype)
+tokenizer = load_tokenizer("omniASR_tokenizer_v1")
+
+pipe = ASRInferencePipeline(model_card=None, model=model, tokenizer=tokenizer, device=device, dtype=dtype)
+texts = pipe.transcribe(["your_audio.flac"], lang=["sna_Latn"])  # LLM: condition on the Shona token
+print(texts)
+```
+
+Audio should be mono 16 kHz (the pipeline resamples if needed); `.transcribe()` also accepts a
+list of raw bytes or `np.int8` arrays. Keep clips under 40 s.
+
+## Checkpoint format
+
+Native **fairseq2** sharded checkpoint (`model/pp_00/tp_00/sdp_00.pt` + `model.yaml`) —
+**not** a `transformers` model, so `AutoModel` will not load it. Load with the
+[`omnilingual_asr`](https://github.com/facebookresearch/omnilingual-asr) / `fairseq2`
+libraries, pointing the ASR recipe's `model.path` at the downloaded `model/` directory
+with `model.family=wav2vec2_llama`, `model.arch=300m`, `tokenizer=omniASR_tokenizer_v1`.
+
+## Citation
+
+This model accompanies the **WAXAL ASR Benchmark** ([arXiv:2606.02375](https://arxiv.org/abs/2606.02375)).
+
+```bibtex
+@article{waxalnet2026,
+  title  = {The WAXAL ASR Benchmark: Fine-Tuned Edge Models Across 19 African Languages},
+  author = {Olufemi, Victor Tolulope and Babatunde, Oreoluwa and Njema, Ramsey and
+             Gbotemi, Bolarinwa and Yen, Wanchi Lucia and Uzodinma, John and
+             Ajayi, Sunday and Williams, Oluwademilade and Moshood, Kausar and
+             Anyaele, Innocent Elendu and Arefaine, Akebert Tesfahunegn and
+             Hunzwi, Candace and Daniel, Wongel Dawit and Namuganga, Emmilly Immaculate and
+             Kadima, Cleophas and Bahizire, Athanase Biluge and Ranaivoson, Onitsiky and
+             Aaron, Emmanuel and Ladislaus, Nicholaus Dismas and Muhammed, Idris and
+             Simenya, Jonathan Enoch and Koome, Martin and Endaylalu, Matewos Tegete and
+             Adeyemo, Peter Ifeoluwa and Birindwa, Hondi Prisca and Eze-Mbey, Ukachi Agnes and
+             Oduro-Yeboah, Yacoba and Aremu, Toluwani and Adjovi, Pericles and
+             Ngueajio, Mikel K and Mitra, Prasenjit},
+  year   = {2026},
+  note   = {arXiv preprint arXiv:2606.02375}
+}
+```
+
+## Acknowledgements
+
+We thank the native-speaker contributors for their language expertise and evaluation support.
+This work was supported by **[Lynguallabs](https://lynguallabs.org/)** (compute, researchers & storage),
+**[Open Token](https://opentoken.global/)** (compute resources), and
+**[CMU Africa](https://www.africa.engineering.cmu.edu/)** (researchers & native speakers).