cahya's picture
Merge branch 'main' of https://huggingface.co/cahya/wav2vec2-large-xlsr-basque into main
0b5ba98
---
language: eu
datasets:
- common_voice
metrics:
- wer
tags:
- audio
- automatic-speech-recognition
- speech
- xlsr-fine-tuning-week
license: apache-2.0
model-index:
- name: XLSR Wav2Vec2 Basque by Cahya
results:
- task:
name: Speech Recognition
type: automatic-speech-recognition
dataset:
name: Common Voice eu
type: common_voice
args: eu
metrics:
- name: Test WER
type: wer
value: 12.44
---
# Wav2Vec2-Large-XLSR-Basque
This is the model for Wav2Vec2-Large-XLSR-Basque, a fine-tuned
[facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53)
model on the [Basque Common Voice dataset](https://huggingface.co/datasets/common_voice).
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
test_dataset = load_dataset("common_voice", "eu", split="test[:2%]")
processor = Wav2Vec2Processor.from_pretrained("cahya-wirawan/wav2vec2-large-xlsr-basque")
model = Wav2Vec2ForCTC.from_pretrained("cahya-wirawan/wav2vec2-large-xlsr-basque")
# Preprocessing the datasets.
# We need to read the aduio files as arrays
def speech_file_to_array_fn(batch):
speech_array, sampling_rate = torchaudio.load(batch["path"])
resampler = torchaudio.transforms.Resample(sampling_rate, 16_000)
batch["speech"] = resampler(speech_array).squeeze().numpy()
return batch
test_dataset = test_dataset.map(speech_file_to_array_fn)
inputs = processor(test_dataset[:2]["speech"], sampling_rate=16_000, return_tensors="pt", padding=True)
with torch.no_grad():
logits = 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[:2]["sentence"])
```
## Evaluation
The model can be evaluated as follows on the Basque test data of Common Voice.
```python
import torch
import torchaudio
from datasets import load_dataset, load_metric
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
import re
test_dataset = load_dataset("common_voice", "eu", split="test")
wer = load_metric("wer")
processor = Wav2Vec2Processor.from_pretrained("cahya-wirawan/wav2vec2-large-xlsr-basque")
model = Wav2Vec2ForCTC.from_pretrained("cahya-wirawan/wav2vec2-large-xlsr-basque")
model.to("cuda")
chars_to_ignore_regex = '[\,\¿\?\.\¡\!\-\;\:\"\“\%\‘\”\\…\’\ː\'\‹\›\`\´\®\—\→]'
# Preprocessing the datasets.
# We need to read the aduio files as arrays
def speech_file_to_array_fn(batch):
batch["sentence"] = re.sub(chars_to_ignore_regex, '', batch["sentence"]).lower()
speech_array, sampling_rate = torchaudio.load(batch["path"])
resampler = torchaudio.transforms.Resample(sampling_rate, 16_000)
batch["speech"] = resampler(speech_array).squeeze().numpy()
return 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):
inputs = processor(batch["speech"], sampling_rate=16_000, return_tensors="pt", padding=True)
with torch.no_grad():
logits = model(inputs.input_values.to("cuda"), attention_mask=inputs.attention_mask.to("cuda")).logits
pred_ids = torch.argmax(logits, dim=-1)
batch["pred_strings"] = processor.batch_decode(pred_ids)
return 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**: 12.44 %
## 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)