---
language: mr
datasets:
- openslr
metrics:
- wer
tags:
- audio
- automatic-speech-recognition
- speech
- xlsr-fine-tuning-week
license: apache-2.0
model-index:
- name: XLSR Wav2Vec2 Large 53 Marathi by Sumedh Khodke
  results:
  - task: 
      name: Speech Recognition
      type: automatic-speech-recognition
    dataset:
      name: OpenSLR mr
      type: openslr
    metrics:
       - name: Test WER
         type: wer
         value: 12.7
---

# Wav2Vec2-Large-XLSR-53-Marathi
Fine-tuned [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on Marathi using the [OpenSLR SLR64](http://openslr.org/64/) dataset. When using this model, make sure that your speech input is sampled at 16kHz. This data contains only female voices, although it works well for male voice too.
## Usage
The model can be used directly without a language model as follows, given that your dataset has Marathi `actual_text` and `path_in_folder` columns:
```python
import torch, torchaudio
from datasets import load_dataset
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor

mr_test_dataset_new = all_data['test']

processor = Wav2Vec2Processor.from_pretrained("sumedh/wav2vec2-large-xlsr-marathi") 
model = Wav2Vec2ForCTC.from_pretrained("sumedh/wav2vec2-large-xlsr-marathi") 

resampler = torchaudio.transforms.Resample(48_000, 16_000) #first arg - input sample, second arg - output sample
# 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_in_folder"])
\tbatch["speech"] = resampler(speech_array).squeeze().numpy()
\treturn batch
mr_test_dataset_new = mr_test_dataset_new.map(speech_file_to_array_fn)
inputs = processor(mr_test_dataset_new["speech"][:5], 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:", mr_test_dataset_new["actual_text"][:5])
```
## Evaluation
Evaluated on 10% of the Marathi data on Open SLR-64.
```python
import re, torch, torchaudio
from datasets import load_dataset, load_metric
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor

mr_test_dataset_new = all_data['test']
wer = load_metric("wer")

processor = Wav2Vec2Processor.from_pretrained("sumedh/wav2vec2-large-xlsr-marathi")
model = Wav2Vec2ForCTC.from_pretrained("sumedh/wav2vec2-large-xlsr-marathi") 
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["actual_text"] = re.sub(chars_to_ignore_regex, '', batch["actual_text"]).lower()
\tspeech_array, sampling_rate = torchaudio.load(batch["path_in_folder"])
\tbatch["speech"] = resampler(speech_array).squeeze().numpy()
\treturn batch
mr_test_dataset_new = mr_test_dataset_new.map(speech_file_to_array_fn)
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
\t\tpred_ids = torch.argmax(logits, dim=-1)
\t\tbatch["pred_strings"] = processor.batch_decode(pred_ids)
\treturn batch
result = mr_test_dataset_new.map(evaluate, batched=True, batch_size=8)
print("WER: {:2f}".format(100 * wer.compute(predictions=result["pred_strings"], references=result["actual_text"])))
```
**WER on the Test Set**: 12.70 %  
## Training
Train-Test ratio was 90:10.
The colab notebook used for training can be found [here](https://colab.research.google.com/drive/1wX46fjExcgU5t3AsWhSPTipWg_aMDg2f?usp=sharing).