---
language: jv
datasets:
- common_voice 
metrics:
- wer
tags:
- audio
- automatic-speech-recognition
- speech
- xlsr-fine-tuning-week
license: apache-2.0
model-index:
- name: XLSR Wav2Vec2 Javanese by cahya
  results:
  - task: 
      name: Speech Recognition
      type: automatic-speech-recognition
    dataset:
      name: OpenSLR High quality TTS data for Javanese
      type: OpenSLR
      args: jv
    metrics:
       - name: Test WER
         type: wer
         value: 19.34
---

# Wav2Vec2-Large-XLSR-Javanese

Fine-tuned [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53)
on the [OpenSLR High quality TTS data for Javanese](https://openslr.org/41/).
When using this model, make sure that your speech input is sampled at 16kHz.

## Usage
The model can be used directly (without a language model) as follows:
```python
import torch
import torchaudio
from datasets import load_dataset
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor

## TODO: load from OpenSLR https://openslr.org/41/
##

processor = Wav2Vec2Processor.from_pretrained("cahya/wav2vec2-large-xlsr-javanese")
model = Wav2Vec2ForCTC.from_pretrained("cahya/wav2vec2-large-xlsr-javanese")

resampler = torchaudio.transforms.Resample(48_000, 16_000)

# Preprocessing the datasets.
# We need to read the aduio files as arrays
def speech_file_to_array_fn(batch):
\tspeech_array, sampling_rate = torchaudio.load(batch["path"])
\tbatch["speech"] = resampler(speech_array).squeeze().numpy()
\treturn batch

test_dataset = test_dataset.map(speech_file_to_array_fn)
inputs = processor(test_dataset["speech"][:2], sampling_rate=16_000, return_tensors="pt", padding=True)

with torch.no_grad():
\tlogits = model(inputs.input_values, attention_mask=inputs.attention_mask).logits

predicted_ids = torch.argmax(logits, dim=-1)

print("Prediction:", processor.batch_decode(predicted_ids))
print("Reference:", test_dataset["sentence"][:2])
```


## Evaluation

The model can be evaluated as follows on the Indonesian test data of Common Voice.

```python
import torch
import torchaudio
from datasets import load_dataset, load_metric
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
import re

## TODO: load from OpenSLR https://openslr.org/41/
wer = load_metric("wer")

processor = Wav2Vec2Processor.from_pretrained("cahya/wav2vec2-large-xlsr-javanese")
model = Wav2Vec2ForCTC.from_pretrained("cahya/wav2vec2-large-xlsr-javanese") 
model.to("cuda")

chars_to_ignore_regex = '[\\,\\?\\.\\!\\-\\;\\:\\"\\“\\%\\‘\\'\\”]'

resampler = torchaudio.transforms.Resample(48_000, 16_000)

# Preprocessing the datasets.
# We need to read the aduio files as arrays
def speech_file_to_array_fn(batch):
\tbatch["sentence"] = re.sub(chars_to_ignore_regex, '', batch["sentence"]).lower()
\tspeech_array, sampling_rate = torchaudio.load(batch["path"])
\tbatch["speech"] = resampler(speech_array).squeeze().numpy()
\treturn batch

test_dataset = test_dataset.map(speech_file_to_array_fn)

# Preprocessing the datasets.
# We need to read the aduio files as arrays
def evaluate(batch):
\tinputs = processor(batch["speech"], sampling_rate=16_000, return_tensors="pt", padding=True)

\twith torch.no_grad():
\t\tlogits = model(inputs.input_values.to("cuda"), attention_mask=inputs.attention_mask.to("cuda")).logits

    pred_ids = torch.argmax(logits, dim=-1)
\tbatch["pred_strings"] = processor.batch_decode(pred_ids)
\treturn batch

result = test_dataset.map(evaluate, batched=True, batch_size=8)

print("WER: {:2f}".format(100 * wer.compute(predictions=result["pred_strings"], references=result["sentence"])))
```

**Test Result**: 25.86 %

## Training

The Common Voice `train`, `validation`, and ... datasets were used for training as well as ... and ...  # TODO

The script used for training can be found [here](https://github.com/cahya-wirawan/indonesian-speech-recognition) 
(will be available soon)