Update README.md

README.md

@@ -1,3 +1,109 @@
----
-license: cc-by-4.0
----
+---
+license: cc-by-4.0
+language:
+- qu
+metrics:
+- cer
+pipeline_tag: automatic-speech-recognition
+datasets:
+- ivangtorre/second_americas_nlp_2022
+tags:
+- audio
+- automatic-speech-recognition
+- speech
+- waikhana
+- xlsr-fine-tuning
+model-index:
+- name: Wav2Vec2 XLSR 300M Quechua Model by M Romero and Ivan G Torre
+  results:
+  - task:
+      name: Speech Recognition
+      type: automatic-speech-recognition
+    dataset:
+      name: Americas NLP 2022 Waikhana
+      type: second_americas_nlp_2022
+      args: Waikhana
+    metrics:
+    - name: Test CER
+      type: cer
+      value: 16.02
+
+---
+
+This model was finetuned from a Wav2vec2.0 XLS-R model: 300M with the Waikhana train parition of the Americas NLP 2022 dataset. This challenge took place during NeurIPSS 2022.
+
+
+
+## Example of usage
+
+The model can be used directly (without a language model) as follows:
+
+```python
+from transformers import Wav2Vec2Processor, Wav2Vec2ForCTC
+import torch
+import torchaudio
+
+# load model and processor
+processor = Wav2Vec2Processor.from_pretrained("ivangtorre/wav2vec2-xlsr-300m-waikhana")
+model = Wav2Vec2ForCTC.from_pretrained("ivangtorre/wav2vec2-xlsr-300m-waikhana")
+
+# Pat to wav file
+pathfile = "/path/to/wavfile"
+
+# Load and normalize the file
+wav, curr_sample_rate = sf.read(pathfile, dtype="float32")
+feats = torch.from_numpy(wav).float()
+with torch.no_grad():
+    feats = F.layer_norm(feats, feats.shape)
+feats = torch.unsqueeze(feats, 0)
+logits = model(feats).logits
+
+# take argmax and decode
+predicted_ids = torch.argmax(logits, dim=-1)
+transcription = processor.batch_decode(predicted_ids)
+print("HF prediction: ", transcription)
+```
+
+
+This code snipnet shows how to Evaluate the wav2vec2-xlsr-300m-waikhana in [Second Americas NLP 2022 Waikhana dev set](https://huggingface.co/datasets/ivangtorre/second_americas_nlp_2022)
+
+```python
+from datasets import load_dataset
+from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
+import torch
+from jiwer import cer
+import torch.nn.functional as F
+from datasets import load_dataset
+import soundfile as sf
+
+americasnlp = load_dataset("ivangtorre/second_americas_nlp_2022", "waikhana", split="dev")
+waikhana = americasnlp.filter(lambda language: language['subset']=='waikhana')
+
+model = Wav2Vec2ForCTC.from_pretrained("ivangtorre/wav2vec2-xlsr-300m-waikhana")
+processor = Wav2Vec2Processor.from_pretrained("ivangtorre/wav2vec2-xlsr-300m-waikhana")
+
+def map_to_pred(batch):
+    wav = batch["audio"][0]["array"]
+    feats = torch.from_numpy(wav).float()
+    feats = F.layer_norm(feats, feats.shape) # Normalization performed during finetuning
+    feats = torch.unsqueeze(feats, 0)
+    logits = model(feats).logits
+    predicted_ids = torch.argmax(logits, dim=-1)
+    batch["transcription"] = processor.batch_decode(predicted_ids)
+    return batch
+
+result = waikhana.map(map_to_pred, batched=True, batch_size=1)
+
+print("CER:", cer(result["source_processed"], result["transcription"]))
+```
+
+## Citation
+
+```bibtex
+@article{romero2024asr,
+  title={ASR advancements for indigenous languages: Quechua, Guarani, Bribri, Kotiria, and Wa'ikhana},
+  author={Romero, Monica and Gomez, Sandra and Torre, Iv{\'a}n G},
+  journal={arXiv preprint arXiv:2404.08368},
+  year={2024}
+}
+```